Technology in the autonomy domain drives a range of vehicle functionality, from ADAS to automation to autonomous driving, and the industry continues to thrive despite challenges from the ongoing semiconductor shortage. In fact, autonomy content value per vehicle will increase substantially by a factor of 3 from today to 2034. However, different regions will experience different changes. European requirements mean more ADAS in every vehicle, while China and the USA are pushing into new technology and use cases with the support of a new generation of suppliers. Still, the challenging macroeconomic environment means struggle, consolidation and shifting expectations and timelines.

The new Automated Driving System (ADS) Regulation EU 2022/1426 covers the approval of driverless shuttles, robotaxis, hub-to-hub commercial applications and automated valet parking, under the small series scheme. It introduces innovative approaches for the definition of safety requirements and validation, blending together different methodologies and following an open-regulation approach. This grants enough flexibility to address the high complexity of ADS systems, but also brings the need to ensure a common interpretation of the regulatory requirements among stakeholders. This presentation will introduce the EU ADS Regulation and the Interpretation Document for its harmonized implementation.

This presentation will provide an overview and analysis of R&I and testing activities in Europe and its Member States and introduce first findings towards the development of a European Framework for testing on public roads regarding testing regulations and harmonization needs. Building on the previous EU-funded Coordination and Support Actions ARCADE and CARTRE, the FAME project develops and maintains tools supporting the coordination and alignment of R&I and testing in Europe. It capitalizes on shared knowledge to improve, across the CCAM stakeholder community, the cooperation, consensus building and data sharing needed for the organisation and evaluation of demonstration in EU.

Safety assurance of Cooperative, Connected, and Automated Mobility (CCAM) systems, is a crucial factor for their successful adoption and deployment in society. This requires a strong safety argumentation, which remains a significant challenge worldwide. Automotive stakeholders generally seem to agree on applying a scenario based approach. But the wide variety of individual initiatives in this field, is tending to silo solutions. The lack of a common approach hampers the large-scale and safe introduction of CCAM systems in our society. It is for these reasons that the SUNRISE project has been initiated, aiming to establish a common Safety Assurance Framework.

CCAV is a joint, UK government policy unit between the Departments for Business and for Transport. This presentation will provide an update on UK government activities supported by £100m programs across R&D/innovation/commercialization, regulation and legislation, and public engagement. These programs build on £440m of joint government/industry investment since 2015.

Automated Driving (AD) mobility services are becoming more and more reality and as a consequence more stakeholders than only the AD developer will work with the definition of conditions for a safe operation: public authorities must understand the definition to give final approval. For that they need a transparent traceability to the conducted tests. Mobility service providers want to know how the system can be applied. They need a mapping to their designated service area. The Operational Design Domain definition should always tell the same story independently how it was transformed. Upcoming standards must and will address these requirements.

In this presentation, Thomas will cover the latest updates on V2x technology developments (ITS-G5, DSRC-Wave, cellular), the latest regulatory and certification situation (EU, US and others) and the latest updates on testing requirements for connected and automated driving. He will also discuss the most recent developments in regional and global interest groups, future challenges and outlook.

Join Andreas Richter from Volkswagen Commercial Vehicles and a supporting team of experts from AVL, dSPACE, Fraunhofer and Robert Bosch to learn how to specify the ODD in a machine-readable and human-interpretable fashion. During the workshop, you will apply state-of-the-art methods compliant with ISO 34503 and consistent with ASAM OpenODD. You will learn how to modularize the specification to enable assembling a single joint specification from components provided by numerous teams and experts. You will understand which use cases and data are required to support ODD evaluation, including discussion of the various technologies and data sources involved. Finally, you will understand how to quantify coverage and determine the impact of specific capabilities on that coverage via ODD comparison methods. ALL DELEGATES ARE WELCOME TO ATTEND AND LISTEN TO THE WORKSHOP CONTENT. HOWEVER, WE ARE LIMITED TO 48 PRACTICAL PARTICIPANTS WHO CAN JOIN IN WITH THE EXERCISES. THESE SPACES WILL BE ALLOCATED ON A FIRST-COME FIRST-SERVED BASIS AND CAN BE BOOKED VIA THE DELEGATE REGISTRATION FORM.

The presentation gives an insight into the basic formula to approve connected and automated vehicles in Europe by highlighting the challenges of the European L4 directive. Furthermore, it shows the release into the market of a member state. It also explains how to deal with scalable compliance over the whole lifecycle from the perspective of a technical inspection company. There is also a focus on the importance of software updates that change the approval. The traceability of those updates must be secured by a continuous homologation process and how the current applicable test criteria must be extended analogously to the complexity of the technology.

Standards are playing an ever increasingly important role in enabling the shift from monolithic to modular toolchains in the industry but they are only part of the picture. We need to take into account additional aspects such as comparability, traceability and consistency, both of data and of the use of the standards themselves, in order to fully enable safe V&V of highly automated driving functions. In this presentation we will show what else we are doing in ASAM and together with other organizations in the industry to achieve this.

Our goal at The Autonomous is to overcome the safety challenges of developing and deploying autonomous vehicles on a global scale. Our Innovation Stream facilitates cooperation across the industry and academic research to develop reference solutions for all these safety challenges (e.g. architecture, AI, regulation). We will give an overview of the ongoing activities, and explain and discuss currently considered solutions (e.g. architecture candidates and KPIs).

Many OEM's and suppliers are investing in the development of automated driving systems from a bandwidth of SAE Level 2 to Level 4. Commercially most relevant are currently Level 2 ADAS systems but in particular premium manufacturers and tech companies are pushing for the market introduction of Level 3 and Level 4 systems. The key question is how regulators will approach the type approval/homologation process for ADAS/AV. Certified cyber-physical testbeds specifically designed for automated driving systems and acknowledged by regulators can play an important role in this process. The presentation will explain the activities of IAMTS in this context.

We propose a standardised method to allow test access to automated vehicles. Borrowing the concept of the standard physical and protocol interface used in on-board diagnostics, this new interface would allow independent test houses and regulation authorities to perform final validation tests uniformly between different production vehicles, with independence and transparency, while protecting OEM intellectual property. Retesting after in-use software updates is facilitated. The ADS is commanded to process synthetic input from a test tool and its planned responses analysed for scenario-based tests; and actual perception performance is assessed using real test track objects. Every vehicle becomes a HIL.

This presentation provides an overview of SAE Standards Activities related to advanced vehicle safety technology. The presentation will focus on advanced driver assistance systems standards, V2X communication standards, on-road automated driving systems standards, cooperative driving automation standards, safety and human factors standards, cybersecurity standards, shared mobility and micromobility standards.

What does agile mean for architects working at the field of ADAS? How is it possible to represent the values of agile in complex automotive projects, and what are the benefits of doing so? In this talk we are going to discuss the differences between classical and agile architecture work, and present various field-proven best practices for communication, change management, technical debt management and customer collaboration for automotive architects.

This presentation will describe how the Autoware Foundation is applying SOAFEE to develop the Autoware Open AD Kit, which is one of the first SOAFEE blueprints. Topics discussed include the architecture of the SOAFEE blueprint implemented in the Autoware Open AD Kit, including the Autoware OSS stack in a containerized/micro-servies architecture, integration with third-party services such as OTA and Mapping, and the cloud-native to edge development and verification framework to enable Software Defined Vehicle (SDV) development and deployment of autonomous driving systems.

With the convergence of high-performance computing, autonomous vehicles (AV) development, AI and hybrid cloud, we'll share the latest technologies, solutions and innovations that make possible data-everywhere and compute-anywhere; develop-once and deploy anytime; cloud-scale elasticity and availability. Kubernetes and containers continue to gain traction for modernizing applications for agile deployments from edge to the core data center to the cloud. We will discuss whether it makes sense for HPC environments or workloads that require huge amounts of data like AV and AI. We will discuss the new Kubernetes container native storage solutions.

The presentation will discuss how to reduce complexity and interdependencies among the many different ECUs, unite applications to unlock new software-enabled functionality and control the software that defines the user experience of vehicles. It will also outline a standard-based approach for interfaces, which improves reuse and creates an open platform for innovation.

The automotive industry was and is still centering around the hardware of vehicles and the corresponding hardware development and life-cycle management. Software, however, is gaining more and more importance in vehicle development and over the entire vehicle lifetime. The vehicle and its value to the customer is increasingly defined by software. This transition towards the so-called software-defined vehicles changes the way to innovate, code, deliver and work together. But how to create ADAS and AD functions for the software defined vehicle?

Automakers have become tech companies with the advent of connected and autonomous vehicles, but lengthy production cycles hinder them from innovating at the speed of the tech industry. In order to keep pace with new technology and emerging trends, auto makers like Toyota, Mazda and Daimler have shifted from traditional development processes to agile, rapid development with Automotive Grade Linux (AGL), an open source software platform for all in-vehicle applications from infotainment to autonomous driving. Dan Cauchy, Executive Director of AGL, will share how 150+ AGL members contribute to the shared software platform, which decreases development times so OEMs and suppliers

When we talk about ADAS/AD it quickly becomes clear that safety is one of the biggest challenges. Jürgen Daunis, CEO at understand.ai explains how Deep Learning / Machine Learning helps to make autonomous or semi-autonomous vehicles safe. He also talks about their effect on time to market and how to achieve the required data quality. understand.ai is on the forefront of ADAS / AD annotation automation. Customers profit from the company’s automotive and machine learning competence to reach the automation levels needed to run large scale validation projects.

Homologation of an autonomous vehicle level 3 and beyond requires a consequent shift-left approach of all development, integration and testing efforts to speed up, scale and make data driven development efforts consistent across all ASPICE levels. Therefore it becomes crucial to have a continuous reusability of knowledge starting from the system description leading to requirements and core architectures up to gaining KPI results. In this presentation Luxoft and AWS will demonstrate how this can be realized at scale covering architectural means as well as digitalized and virtualized and therefore scalable testing of digital twins

Datasets are used widely for the validation and training of perception stack in self-driving. The needed effort for annotation is directly proportional to the amount of recorded data. This becomes quite expensive and time-consuming when it comes to having big chunks of recorded data. Brightskies has built an AI engine called BrightAnotate that is capable of automating ground truth generation for road topology, infrastructure, static environment, and traffic participants fusing multiple sensors like LIDAR, 360 cameras, radars, and GPS. Combined with a concrete annotation process, Brightskies has successfully reduced the manual effort needed by 50% and hence improve overall productivity.

As the industry searches for sustainable paths to deploy increasing levels of driving automation, the choice of compute platform becomes central to the business strategy of automakers and suppliers. While meeting diverse performance, power and cost requirements can be challenging, similar consideration must be given to managing software lifecycle investment across vehicle platforms. This presentation explores upcoming vehicle compute architectures trends and some of the foundational technologies, across hardware and software, required to deliver scalable compute platforms.

ADAS/AD requires special functionality for the middleware: high data volume must be handled in an efficient way, functional safety up to ASIL D has to be fulfilled, and the system must behave deterministically in the real and virtual world to enable efficient development and reliable virtual test and validation. All this is not addressed sufficiently in current middleware solutions. We will show these specific ADAS and AD requirements and discuss ways how to address the above mentioned challenges in the development.

Autonomous vehicles leverage artificial intelligence commonly in their perception functionality, i.e. for camera-based computer vision, radar or lidar postprocessing. There, AI benefits the quality of the perception in that the detection and classification of environmental aspects is more precise than with classical algorithms from pre-deep learning times. Similar benefits may also be yielded with AI for planning and control. However, this is not very common up to now. We present a variety of concepts and results of our work on how AI can be used for planning and control to increase passenger comfort and energy-efficiency while considering safety.

At the innovation line driverless, we’re operating prototypes to understand the challenges and impact of AD. At the core of the AD system in our vehicles is a dynamic grid based fusion. It provides free space and generic objects to our planning stack in prototype vehicles to ensure the planning of a safe trajectory also when encountering unknown objects. We show examples of the system behavior in challenging perception and planning scenarios. To further push the performance and reliability of the perception, we show how AI based methods fit into the inputs and outputs of our fusion system and how we integrate these into our data loop. Based on our experience of the demand for high-quality sensor data we developed the concept of the CoSAr, which is the next step in the industrialization of L4 vehicles by providing a multi modal sensor array solving the challenges of calibration, integration and maintenance also when the number of vehicles is scaled up.

In the future, simulations will be of considerable importance for quantitative testing and validation of future automated driving functions. However, in order for these to be able to fully and reliably replace real driving situation, it is necessary to be able to map this real traffic virtually - especially with regard to critical situations. The developed incident detector application for weak sensor signals, automatically evaluates and labels critical situations in which sensors and their associated object detection algorithms could have difficulties. So the incident detector application has on the one hand a trained AI to detected and evaluate this situations automatically. On the other hand you have an edge case scenario database where you can easily browse and find relevant cases for your development and validation job. e.g. overexposure after exiting a tunnel. In a next step the image information for e.g. the overexposure after a tunnel can be used in the simulation with many variants in multiple real-time for the development and validation of new automated driving functions.

Validating perception through physics based sensor models, demonstrating the safety of AD/ADAS systems or generating virtual datasets for machine learning challenges traditional desktop based simulation by requiring to run thousands or millions of simulations. SCANeR Cloud, developed by AVSimulation and powered by Amazon Web Services, is a Web-based massive computing platform that allows to prepare, run and analyze millions of SCANeR simulations in parallel including virtual scenario, traffic, sensors, vehicle dynamics, etc.

The race of the automotive manufacturers to bring new technologies into vehicles is fueled by tighter emission standards, stricter safety requirements and higher levels of driving automation. Racetrack simulation is becoming increasingly important, as it provides a safe, cost-effective, and efficient means of evaluating vehicle performance. In this presentation, we will showcase our solution for a modular simulation environment that combines a high-fidelity track environment and a full vehicle model. Our approach uses an optimized Model Predictive Control (MPC) and includes a method for optimizing the trajectory and speed profile to generate a realistic driving profile with best lap performances.

Proof of functional safety for automated driving must be based on evidence from virtual validation, as the necessary evidence cannot be generated with test drives solely. A challenge is generating relevant test cases, as there are infinite simulation scenarios and variations. RevoAI and Fraunhofer IESE present a solution to find relevant and previously unknown scenarios efficiently. They have developed an AI that learns the behavior of the test subject of an automated driving function in its environment during simulation and generates challenging and critical scenarios for the test subject. This approach allows an efficient safety evaluation of the test subject.

Developing L2-L5 autonomous vehicles requires significant resources – time & infrastructure. With pressure to reduce ADAS/AD development cycles, and limited budgets, sizing the right amount if infrastructure – storage, networking, CPU & GPU compute becomes critical. You must also solve sensor data management challenges across the entire data lifecycle – from edge-to-core-to cloud. In this session we demonstrate how to size your infrastructure for HiL and SiL testing, while considering future expansion, using hardware integrated with industry-standard software and applications. The focus will be on real-world variables that must be considered to optimize infrastructure for cost, performance and flexibility.

The SOTIF (Safety Of The Intended Functionality) standard has brought along a shift in the safety case, with a vehicle manufacturer carrying an increased liability. Siemens has developed a proprietary methodology to systematically generate unsafe-unknown scenarios for a specific ODD and (optionally) recorded data. By adopting the Critical Scenario Creation process, OEM’s and AV suppliers have a methodology to automatically generate unknow-unsafe scenarios. Additionally, cities and traffic planners can use the tool to identify problem areas and consequently design and implement infrastructure that supports the safest operations, thus minimizing risks in the urban environment.

In order to accelerate the development of AV and ADAS systems, often the vehicle's Software (SiL) or Hardware (HiL) is tested in a simulated 3D Environment. The information from simulated sensors such as HD cameras, LiDAR and Radar is transferred to the vehicle's computer. The AI Driver observes the vehicle's surroundings whilst driving in a simulated environment. In order to maximize its potential benefits, the 3D Environment need be highly realistic and accurate. Various methods are used to maximize the realism of the Digital Twins of streets and terrain. These methods and challenges will be addressed in this presentation.

Infrastructure support could make automated driving much easier in many scenarios and accelerate the safe introduction of automated driving significantly. Though the provided information must comply certain requirements in from and content to become safe and reliable enough, to deliver real added value. The proposed presentation aspires to demonstrate how a Digital Twin must be setup for being a reliable and advantageous Decision Support Platform in Cooperative, Connected, Automated Mobility (CCAM) to overcome some of the conceptual ODD limits in autonomous driving. The presentation sums up the results of several research projects like DIGEST and COPE (Collective Perception).

As urban populations continue to grow and more people choose to bike or use micro mobility solutions, traffic in cities is becoming increasingly complex. Today, 40% of all fatal traffic accidents occur in urban areas, according to the EU Commission. In response to this issue, new players are emerging and driving the development of the next generation of smart, safe, and energy-efficient vehicles. However, the current ADAS technologies on the market lack sufficiently fast detection and reaction capabilities to ensure the safety for vulnerable pedestrians and cyclists when traffic is dense and varied.

Nemo by Ridecell is a data platform used to extract interesting scenarios from vehicle data that helps ADAS and AV developers accelerate testing and validation and scenario coverage / ODD quantification. We use our proprietary scenario description language, called Ridecell Scenario Language (RSL), to describe scenarios of interest. RSL is also used to search for interesting scenarios in graph based representations of driving events. We are currently working with large OEM and tier-1s processing millions of kms of real world production data per day on the platform. I will discuss learnings from our deployments and advantages of our approach.

In recent years, the open ASAM standards (such as OpenDRIVE and OpenSCENARIO, Open Simulation Interface) has emerged as big players in the AD/ADAS simulation community. To build a common understanding on how the standards should be interpreted, open-source tools has one big advantage: They are open for everyone. esmini and scenariogeneration are examples of such tools, stimulating the spread, usage and harmonization of the standards. This presentation will introduce these tools and show a few use cases.

Proving the safety of highly complex, automated systems used in Connected and Autonomous Mobility (CAM) domain requires a paradigm shift in current test methodologies to enable more collaboration and synergies among the stakeholders. Virtual testing by use of simulation tools has gained immense popularity in the CAM domain. Although there several simulation tools available in the market the existing tools are extremely fragmented and complex. Distributed co-simulation can be used to derive optimised combinations of simulation tools, leveraging their strengths and attenuating their weaknesses. KAN-Do is a tool-agnostic, distributed co-simulation platform providing a secure, data-centric, end-to-end ADS development framework.

ADAS and AV functionalities built around perception components must be tested rigorously and early in development to achieve safety and reliability across adverse conditions and scenarios. However, challenges remain in assuring safe performance. This talk will discuss solutions to these challenges derived from our joint work with an OEM on safety evaluation of a traffic sign classifier for an ADAS highway pilot feature. We will show how the testing tool aidkit facilitated ODD-based performance analyses for perception models based on diverse data augmentation techniques. Additionally, providing insights regarding the integration of aidkit into safety validation processes, as well as method validity experiments for various automotive use cases.

What is the difference between ADAS and Autonomous driving ? The answer is… safety design and demonstration level. The ISO standards define the objectives and the global methodology and regulation defines a process based on 3 pillars: proving ground testing, open road driving and virtual simulation. But the way to design a safe Autonomous Driving System and demonstrate the achievement of the safety goals is not well defined nor standardized, especially for highest levels of automation. Some fragmented elements exist and have been applied for first applications, in general by limiting the ODD, including redundancy in hardware and software, using data collection and re-simulation and driving massive mileage with dozens of fleet vehicles. The use of virtual testing with massive simulation guided by a scenario approach will be key to be able to put on the market Autonomous Driving System with extended capabilities. The presentation proposes a global process which articulates the different methods with a highlight on virtual validation.

This presentation will give an overview of results of the French public funded project named SAM for Safety and acceptance of Autonomous Mobility. These results where presented to French authorities to bring answers to the questions posed by the establishment of French new mobilities law published in 2022 and international regulations concerning automated vehicles and automated road transport systems. The approach starts by stating that such systems shall be designed and validated considering GAME (French acronym for Globally At Least Equivalent) safety principle and nominal scenarios from system design, scenarios from risk analysis, scenarios from accident analysis, and scenarios from real-word driving.

This presentation will discuss the limits of physical tests (large test matrix, destructive tests, limits of test resources, etc.) It will look at VIL, DIL, SIL applications, VT & homologation (GSR2 / R152 etc.) It will offer perspectives on adverse weather testing, AI evaluation and the introduction of V2X.. Lastly it will look at the Dassault Systems/AVS collaboration.

French publicly funded research projects are ongoing on the topics of safety, safety demonstration, numerical modelling and simulation tools, and scenario-based automated system design & validation, and homologation. These projects allow the French automated mobility eco-system to build the French legal framework for automated vehicle homologation and for authorization before commercial services of automated road transport systems. Among these tools, the scenarios library play a key role. Based on research assets, developed for 6 years now, some OEMs, decided to industrialize a scenarios library. This initiative, supported by PFA will be presented in depth, its name is ADScene project.

Major updates of rating and regulation systems for vehicle approval on roads is ongoing. These are including updates on existing ADAS systems but are also adding new systems to be implemented on road vehicles. Those new procedures increase drastically the number of regulation and normative tests to be performed as well as the tuning tests by the manufacturers. To deal with this major fact, and still reduce development timings, current testing and simulation solutions could evolve. This discussion will show some aspects of the tests on track and VIL usage that will help dealing with time reduction. This presentation will address how best to set up a VIL method to fit these requirements

Model-based systems engineering is a methodology that focuses on creating, maintaining, and exploiting models as the primary means of systems analysis and engineering collaboration between different disciplines. We apply these principles to the triumvirate of safety by design, safety by verification & validation and the delivery of incremental safety cases by combining simulation with analysis and scenario at scale for L3 autonomous functions. The presentation demonstrates Ansys’s commitment to safety, we will introduce customer use cases which illustrate proven applications of our approach to help bring safe and reliable L3 functions to the market.

In this presentation, we will oversee the main challenges ahead for the validation & verification of autonomous driving systems. In the top-down phase of the development, we start with a list of the functionality and the global ODD objective of the system. This phase, qualified as “design”, helps in refining and defining the intended use of the system and to divide it into unitary functions easier to design and to understand. In the bottom-up phase, there is no unified and formal solution to prove the completeness of the validation process. Therefore, we will give an insight into existing methodologies, frameworks and finally show the status of regulation of ADS at European and International level.

As already widely presented, AD/ADAS systems validation is a complex task, and the current trend is to perform and combine various validation activities: field testing, bench testing, simulation, at different levels of system integration: model based, software based, hardware based, vehicle in the loop, driving simulation … The presentation will be centered on simulation activities for validation, addressing some trends at Renault: shift-left approach regarding the development cycle, where simulation can offer earlier and pre-validation results thanks to MBSE implementation. Building appropriate and automated simulation platforms addressing the different use-cases and validation needs. A perspective of some hot topics will also be presented: Impact of new vehicle architectures (namely Software Defined Vehicle), Contribution of simulation to AD/ADAS approval & Inter-systems validation

Automated driving complexity has grown exponentially requiring reinventing development and validation methodologies to address end-to-end system of systems architecture and deliver certification driven simulation and Virtual Testing. The capability to model and simulate requirements and architectures in early phase of a vehicle program enables early error detection and convergence on on best trade-off in term for safety and operational design domain (ODD). Massive simulation, generated from multiple scenarios and correlated with real on-track data, eases the validation plan coverage and the optimization of physical road tests. End-to-end change management and traceability accelerate standards and regulations compliance.

The presentation describes a new approach to safety verification and validation of ADSes. The approach relies utilizes ASAM OpenSCENARIO® 2.0.0, coupled with scenario based coverage driven validation . This approach changes the way the automotive industry is conducting verification and validation of automated driving systems. It also demonstrates an advanced method for safety verification that allows tackling the infinite space of scenarios. Several use cases will be presented, with an emphasis on how the new approach enables a step function in ensuring the correctness and safety of automated driving systems.

In the last 1.5 years, Torc Robotics and TNO have implemented large scale scenario detection using TNO’s StreetWise pipeline. During this work, the need was felt to develop a more structural approach for defining scenario categories. Good coverage of real-world driving with scenario categories is an essential part of developing Level 4 autonomous vehicle technology at scale. This presentation will present various approaches to identifying scenario categories, such as knowledge-based approaches that look at vehicle combinations, unsupervised methods, tag-based approaches, ADS-focused approaches and discuss advantages and disadvantages. The right set of scenario categories depends on the purpose and use case.

We present a localization solution for autonomous vehicles that merges heterogeneous observations (sign detection, dashed markings, buildings and GPS). Based on a particle filter, the localization solution takes into account the asynchronicity of the data in order to estimate the most probable state of the vehicle. The general approach, then the implementation details concerning the selected types of fixed references are presented. Finally, an experimental work at the scale of the city of Versailles has been performed. The first results presented are very encouraging. The fusion manages to switch from one method to another or combine several approaches at the same time.

This talk will focus on revealing the myths and the reality associated with realising a scalable safety assurance process for automated driving systems. While there is widespread acknowledgement for the need for an Operational Design Domain (ODD) based safety assurance framework, various aspects of the framework are either misrepresented or not considered at all. This talk will focus on some of these aspects, uncovering myths (and the reality) associated with simulation-based testing, ODD definition, and wider scenario metrics. While simulation-based testing remains a key component of the framework, more focus needs to shift in understanding ODD definitions and validity of simulation to give confidence in the output of the testing process.

There are several challenges that need to be focused once emphasizing on the scenario-based testing of autonomous vehicle in Malaysian road and traffic environment. Hence, a safety assessment in virtual platform for autonomous vehicle is required as part of the homologation process to further enhance the deployment of autonomous vehicles in developing countries, such as Malaysia. The platform can be used as the first layer of testing procedure for autonomous vehicles using various type of scenarios and testing standards before focusing on physical testing.

Undoubtedly, ADAS functions require scenario-based testing to evaluate their closed-loop behavior. This holds true from early development (MiL) to HiL testing resulting in manifold test execution environments. The creation of scenarios and test-cases is a resource consuming task, making it highly desirable to re-use them throughout all development phases and in all execution environments. Open standards such as OpenSCENARIO build a strong basis for this requirement. This presentation sheds light on how to describe and execute tests and scenarios and evaluate the results using standards with the goals of seamless re-usability and automation in mind.

ADAS has the potential to reduce the occurrence and severity of vehicle collisions. However, a gap exists between real-life winter environments and published ADAS test protocols for which assessments are performed under ideal conditions. Since 2016, PMG Technologies and Transport Canada have been performing track testing to evaluate the effect of winter conditions on ADAS. Winter-specific test parameters include roads, targets, and vehicle sensors covered with ice or snow, pedestrian targets dressed in winter clothing, and snowfall intensity. This presentation describes the challenges of winter track testing, the required modifications to methodologies and test equipment, and the test results.

The presentation focuses on the different approaches followed in the testing and validation of ADAS and AD systems and its features. It covers mainly how different test types are selected, organized and executed. How to make sure that right testing is performed virtually and physically in real world. Also discusses about the possibility of reducing the deviations while using virtual environments for testing.

Starting 2014, over the course of the past 9 years EasyMile has built up a visible and respected project portfolio of autonomous passenger transportation across Europe. Shifting from deployments with a non-type-approvable vehicle platform based on local exemption permits with the requirement for on-board safety attendants, to today's fully driverless and remotely supervised deployments on open roads is a major step forward for the AV industry in Europe. The presentation will focus on key learnings of the past years and critically reflect which elements are needed 2023 onwards to allow a large-scale adoption of AV technology in Europe.

Adastec Corp. deployed the first real-life pilot of a full-size, electric, Level 4 automated transit bus in the United States (Michigan) and Europe (Norway). Both buses are operating on public roads and carry passengers on a daily basis. Adastec Corp. tells the story of the future of electric and autonomous transit buses, the road to receiving NHTSA/TUV Nord approval, collaboration efforts with municipalities, bus OEMs, bus operators, federal, state and local governments, and the journey to becoming the first and currently only autonomous bus software provider in the United States and Europe.

Road markings are essential to ensure good driving conditions. Traditionally, the characterization of the road marking has been standardized by the visibility of the human drivers. However, AV technology relies on a machine perception of road markings. This raises the question whether the current standards and models need to be adapted to enable a sufficient quality for detection by AV's. This presentation summarizes the result of two projects which investigate what affects the detection, and the road authorities' need for maintenance of the road infrastructure to meet the demands of the future with automated driving.

With the development of vehicle automation, more and more complex situations have to be examined. So far, simpler tests with lower speeds from target's perspective have been typical, which corresponds to road or city-like situations. However, driving assistants operating in L3 highway environments require much longer, more complicated scenarios with multiple actors and higher speeds. For effective testing, it is essential to rely on simulations. AVL ZalaZone's engineering team started implementing simulation-based tests according to industrial needs. This presentation describes the inherent advantages and challenges and shows the process of test executions derived from the digital-twin.

A data-driven methodology that aims to identify unknown driving situations which trigger unexpected behavior for ADAS/AD functionalities over time in order to minimize unsafe situations. This is done by analyzing the collected data for criticality KPIs and creating specific test-scenarios that are used for virtual testing, and on our test-track by our main customer AVL. The data collection approach fuses lidars and cameras, and is currently deployed over 15+ locations around Austria. It tackles the fidelity and representativeness of scenario-based testing and enables a gap analysis between human-drivers and autonomous vehicles through a detailed driving behavior analysis over different conditions.

Over the air testing of automotive radar has multiple challenges mainly due to mmWave frequencies. To name a few, selecting the right chamber, setup cost, floor space requirement, accuracy, repeatability and uncertainty of the test system. With 4D imaging radars entering mainstream, the challenges are becoming more evident. This presentation is aimed to discuss different OTA test methods and their tradeoffs. A novel OTA path loss calibration will be introduced to address some of the aforementioned challenges.

In the scope of rapid development and huge data, a better method to represent the LIDAR ECU perception stack through simulation is needed. This helps different stakeholders to use simulation for sensor fusion and vehicle architecture testing ensuring real time capabilities. In this presentation, the LIDAR phenomenological sensor model will be presented.

Kautex shows how sophisticated testing capabilities help to test customer specific sensor-sets with actual vehicle design under dynamic conditions. LiDAR cleaning performance is tested in our dynamic test bench under different vehicle speeds e.g. 0km/h, 120km/h and with different nozzles and cleaning parameters. Depending on the customer requirements, the system is optimized for the most critical speed; and in case of an intelligent cleaning system, the parameters are adjusted while driving.

An overview of autonomous navigation research and development activities at Technology Innovation Hub on Autonomous Navigation (TiHAN) at IIT Hyderabad, will be introduced. LiDAR odometry and Mapping algorithm creates an HD map of the environment. On the map, the vehicle is localized in real time, which helps in navigation and path planning of self driving vehicles. Efficient algorithms are developed for obstacle avoidance during navigation. HD map based navigation on autonomous vehicles is suitable for controlled environments and GPS denies scenarios. The proposed HD map based navigation algorithm is implemented in real time on an AV, tested and validated.

Today, automotive radar is one of the leading technologies for advanced driver assistance systems (ADAS) due to its unique ability to operate independently of weather and light conditions. Millions of cars have already been equipped with radar sensors, helping to increase the drivers’ comfort and safety. Heading toward autonomous driving (AD) requires robust and highly accurate environment sensing. For this purpose, the potential of radar for 4D imaging and the use of radar networks will be further explored, which requires testing and simulation of the entire ADAS/AD system under 'real' environmental conditions already in an early stage of development.

Testing of new automotive sensors is usually left until the end of the design process, but the increasing complexity of sensors and the development of sensor fusion software demands a shift to continuous testing and feedback. Doing this will shorten time to market, save money in expensive road testing and ultimately deliver a product that exceeds your engineering use case. In this presentation Raphael will explain why the only way to address these challenges is through a new approach to vehicle electronics with a design programme of continuous lab testing integrated into the development programme.

Projects for ADAS/AD platforms are increasingly moving to a data-centric architecture. At the same time increasing hacker attacks show the vulnerability of companies R&D. In addition, data protection regulations are implemented stronger and more consistently. These are good reasons to keep the data trustworthy - including correctness, completeness, encryption. At the same time we need high performance due to 100Gbit/s plus sensor bandwidth. This presentation shows a solution and architecture with software function blocks on mobile edge units for trustworthy and performant recording of the data during the test-drive from the sensor to the cloud.

Deep learning AI has triggered the first revolution in autonomous driving. The breakthrough on the mass market is currently prevented by the fact that this technology is inherently uncertain. Modern deep learning models hit a reliability limit of about 95 percent because of the unknowns. Given an unknown input, the output of a deep learning model is undefined. We have developed a very efficient new methodology to detect the unknowns based on uncertainty in real-time. With our approach, we can develop safety proofs and enable certification/homologation of autonomous vehicles without driving billions of kilometers.

There are several key factors that can contribute to the success of an autonomous development project using simulation. Besides an accurate and detailed virtual environment you should have a flexible and customizable simulation environment. It should allow developers to easily change and test different aspects of the environment and the autonomous system's behavior. This can include things like the algorithms used for decision-making and path planning, physical parameters of the vehicle or robot as well as sensor parameters. It is crucial for developers to not only have access to the data but have a higher control over it.

The Hi-Drive project is pushing automated driving further towards High automation. The feasibility of high-level automation is tested in different conditions across Europe from south to north free from earlier narrow Operational Design Domain (ODD) characterising SAE L2-L3 automation. This presentation about Logging tools recommendations for HiDrive EU-funded project responds to variable research questions as formulated in the project methodology and further on via a concrete signal list as defined for the evaluation needs. Furthermore, even broader logging needs were considered when making recommendations for data loggers. Several reference loggers were defined and recommended which met the aforementioned needs.

ADAS/AV feature development is a complex process that starts with ingesting, labeling, and cataloging hundreds of petabytes of data. Searching through the data repository for scenes relevant to specific models being developed and trained is a tedious, time-consuming process. In this session, we’ll introduce how the AWS Autonomous Driving Data Framework (ADDF) helps accelerate the development process. We will show how the scene detection module can be used to run scene analytics for lane detection and topic synchronization, and finally demonstrate how the visualization module can be leveraged to seamlessly stream and visualize stored ROS bag files.

The physical world is complex and highly dynamic, with sudden changes and unforeseen anomalies. Therefore, it is essential to establish a systematic way to prepare autonomous systems for safe operation in the physical world. In this presentation, we give insights into our novel AI Scenario Engine that helps customers deploy autonomous systems significantly faster and with less cost. Core of this platform is a world-class computer vision software that extracts traffic scenarios from monocular cameras in a highly accurate and fully automated way.

To enable innovation, software-defined vehicle architectures must provide flexibility, scalability, compatibility and upgradability on different platform components. The new vehicle development paradigm requires safety critical software to meet the safety requirements set forth in the Functional Safety (FuSa) standard for ISO 26262 certification. Platform-independent solutions can help OEMs to optimize the path to safety while reducing associated risk and cost. In addition, choosing the right Business Model and liability allocations will help ensure long-term success. This session will highlight challenges and options for a solid safety architectural strategy for production-grade vehicles.

How can we enforce policies that can be interpreted and shared by machines? Existing regulations have been drafted by humans for humans. Translating laws into machine-interpretable code will allow us to govern the behavior of autonomous systems in a policy-compliant manner, while adapting them as laws change. Under Flying Forward, a Horizon Europe project, VERSES and its partners developed a geospatial infrastructure using IEEE’s Spatial Web standards to automatically enforce rules and policies in Urban Air Mobility. Our solution allows current laws to be parsed into machine-readable models and interpreted programmatically, reducing human interpretation and automating their enforcement and auditability.

Advanced AVs for SAE L3/L4 functions will lead to a new understanding of the operation phase in the product lifecycle. The regulation like the EU Implementing Act and the German L4 act (AFGBV) request also a continuous field surveillance, the handling of critical E/E faults and SW updates during operation. This is required to enhance the ODD during operation, offering Function on demands (FoD) and to reduce downtime by a SW stack E/E malfunction. The handling of incident/accidents, caused by E/E malfunctions, will require a minimal “time-of-action” to analyse the root cause, supported by the visualization of the effect chain.

State-of-the-art general-purpose AI models (such as GPT-3 by OpenAI) do not deal well with highly technical language or regulation. They do not have access to the latest information about products (specific terminology) and are known to produce unreliable answers. To solve this problem, deep-learning models need to be trained specifically on technical content, such as manuals and standards. These models power software solutions generating substantial added-value to OEMs, Vendors and Suppliers in the Autonomous space.

OEMs require hundreds of drivers and hundreds of hours of data to validate Driver and Occupant Monitoring Systems (DMS/OMS). In real-world data collection campaigns this is very expensive and time consuming. Luxoft has developed a DMS/OMS Virtual Validation Toolchain to reduce both, effort and costs significantly. A critical point is to convince OEMs that the used virtual videos are photo realistic. This presentation will first show, how to answer this question and shows the advantages of virtual validation for interior sensing. Finally, a demonstration of the DMS/OMS Virtual Validation Toolchain will be given.

Semi-automated vehicles on SAE L3 are now entering the market and driver monitoring to observe take-over availability is needed for vehicle certification. However, driver state classification based on AI methods for drowsiness today lacks reliability in accuracy, robustness and predictability. Here we present the results of an extensive driver simulator study that shows top performance in drowsiness classification. Ninety-two volunteer drivers were tested in alert and drowsy state as well as in manual and automated driving state. We also present how to access to this database, which is unique in terms of quantity and quality of the data.

One of the most important factors for the successful deployment and public acceptance of AVs is the comfort (physical and psychological) of their users or discomfort due to motion sickness. It is influenced by vehicle controller settings and driving style, lack of control and trust, low situational awareness, and potential engagement of users in non-driving related tasks (NDRTs). This presentation will explain various factors that affect user (dis)comfort in AVs and provides a state- of the art review of technology and methods for objective assessment of motion sickness, as well as some potential approaches to actively mitigate it.

In the 2020s datalogging products became available for even the most demanding high data rates. Now that camera megapixel and resulting data rates no longer double every two years, new future requirements will evolve into challenges for today’s measurement-oriented datalogging products. The datalogger as an edge device already is an ongoing development, with stakeholders all over the place who want a piece of the action in both the vehicle platforms and the cloud. Then there is an almost untouched topic: safety and security in vehicle networks.

In this lecture, we will discuss the Cognata Datalab system, an interactive and automatic system specifically designed for training and evaluating the performance of autonomous systems such as AV and ADAS. We will explore how the system uses synthetic data and a simulation system to train and fine-tune neural networks until optimal performance is achieved, resulting in more robust and reliable autonomous systems.

In order to make automated driving a reality for people all over the world, car manufacturers have to develop automated driving features and prove their safety and security. The collection and analytics of mass data produced by vehicles is a key factor. This presentation will explain opportunities and challenges for the use of mass vehicle data during all stages of development on an exemplary automated driving feature. Showcases of practical use within a big-loop ecosystem with dashboards and key performance indicators are presented. These methods facilitate engineers to continuously evaluate and improve their function over the entire lifecycle.

One of the biggest challenges facing the AV industry is the collection of high-quality, safetycritical data to train and develop vehicle sensor systems. In this presentation, rFpro will provide technical insights into its new ray tracing rendering simulation technology, that, for the first time, accurately replicates what sensors observe, enabling the use of simulated data to train perception systems. It will reveal how a technical partnership with Sony Semiconductor Solutions, is integrating sensor models into rFpro’s new simulation technology. Together, this enables development to begin, before physical sensor prototypes have even been built.

Test fleets generate terabytes of raw vehicle data that must be easily offloaded with minimum vehicle stop time. Microsoft and Xylon enable test drivers to bring harvested data to globally accessible computing clouds in the shortest-possible time. Without hardware modifications, Xylon’s logiRECORDER Automotive HIL Video Logger enables direct plug-in of Microsoft Azure SSDs and full-speed recording of encrypted data. Filled SSDs can be exchanged within seconds and shipped to the nearest Azure region center. The presented solution completely eliminates the time-consuming and costly process of copying data from the logger’s storage to the storage media suitable for cloud services.

The validation of ADAS/AD functions requires the use of different test methods to holistically validate the entire application. Separate tools and frameworks today make efficient validation difficult. In this presentation, dSPACE proposes an integrated framework with global data management to ensure efficient collaboration between distributed teams and overcome common weaknesses. Combining the test methods scenario-based testing and data replay testing in this framework enables back-to-back testing with real-world data and synthetic scenarios generated from this real data. This increases the validity of test results, saves time and costs, and enables additional use cases such as model validation.

Limitations in physical testing made virtual validation of highly automated driving functions mandatory. The model-based development approach is accompanied by deep supply chains of simulation artifacts. For the composition of application-related digital twins, the availability of qualitatively suitable data from different sources is crucial, which is becoming an increasing challenge. The availability of data is impacted by defining clear requirements in a trusted data ecosystem. For this reason, around 20 well-known industrial companies and research institutions are exploring solutions in the collaboration project GAIA-X4PLC-AAD for topics such as access management, data sovereignty, data quality assessment and use of data spaces.

Log data is one of the most essential building blocks of autonomous systems development. However, due to the costs and risks involved in real-world testing, autonomy programs must collect and manage their log data effectively. In this presentation, we will discuss the journey of a log file, starting with log collection and exploration to storage and archival. We will also discuss the technical building blocks, ideal workflows and cost-management strategies of an expansive log management process.

The MOSAR platform offers a scenario based methodology and its associated tools designed for research activities on the safety of automated driving systems. It notably includes a scenario manager and a safety relevant scenario library which as been successfully supplied with driving data, accidentology studies as well as expertise scenarios and use-cases. It is the result of several projects involving the IRT SystemX (Institute of Research and Technology) and leading French automotive and public transportation players. We propose a presentation of the MOSAR framework and its surrounding concepts.

In recent years, the amount and complexity of ADAS/AV tests increased, this drives higher demand for proving ground area usage. Furthermore, the validation of simulation results creates the need to map these scenarios into the real world seamlessly. To manage these challenges integrated software solutions need to be developed, that increase the efficiency of time and area usage, not only on the test track but also in the office. In addition, these solutions must add further safety features to enable safe testing.

To test ADAS Domain Controller function, the developer need the HIL system with sensor emulation method.In this presentation, we will show our latest HILS technologies with followed topics. - Video Data Direct Injection with RDMA(Remote Direct Memory Access) -Radar Raw Data Direct injection -Ultra Sonic Object Simulation -GNSS Target Simulation The HIL has to be combied all environment simulation method with raw data or object data emulation.

With the enacted Autonomous Driving Act in 2021, Germany created the world’s first legal framework for Level 4 vehicles on public roads. Where do we stay now? What is the current legal framework, also beyond Germany? With the disruptive entrance of new players in the market, what challenges does it impose on our concepts of liability and warranty? The presentation will also give an outline of the remaining challenges regarding the issues of data privacy and data security.

Now that we have removed the human from the vehicle, has the AV become safer in traffic? No. We have removed the human and not the human element. But wait, is removing the human the solution? No. In fact, the opposite is true: the more the merrier. In my multi-pillared approach to continuously improving and deploying safe and trustworthy autonomous vehicles, inclusion is key. Data, verification and personas need to be weighed equally. Humans in our different roles (as engineers, as drivers, as traffic inspectors, etc) are to be modeled using AI in order to understand and prove the safety of AVs.

The SAFE-UP project demonstrates how V2X connectivity can improve the current ADAS systems (e.g. AEB) on selected scenarios. The Connected-AEB function developed within the project leverages SOTA technology and innovative techniques to feed the data fusion system with additional perception information from external trusted sources, which provides wider and better visibility of the objects around even with NLOS (Non Line Of Sight) conditions. This presentation will include the methodology followed to develop such system, which could be of reference for incoming V2X-enabled ADAS systems.

Testing of an ML- and AI-based perception system shall be realized as a part of the safety assurance of ADAS and AD. We speculate a safety-aware perception system to be validated needs the knowledge of an ideal result. This ideal result, a validation data set, shall contain all the valuable information about the surrounding traffic and environment the perception system should perceive. Therefore we need to address it as a part of the perception system safety case. The presentation will provide sufficient argumentation on the safety aspects a validation data set must fulfill to validate a perception system at different stages.

Presentation includes an overview of the generic road weather forecasting process and how that process must be tailored to the special use cases of AD/ADAS. The pros and cons of vehicle sensors and road weather forecasts are reviewed in the AD/ADAS context, and the benefit of using fleet sensor data to improve road forecasts is quantified. Finally, a brief tour of weather phenomena relevant to AD/ADAS is given, along with the predictability of each phenomenon.

The presentation will give an overview of various safety assessment and residual risk quantification methods for autonomous vehicle development and homologation and requirements according to SOTIF standard and UNE157 ALKS regulation. It will demonstrate the required workflow from real-world data to scenarios to safety and risk assessment and visualization. Furthermore, it will present in detail a unique approach that AVL has developed together with a partner to evaluate risk levels from scenario data and real-world statistics within the AVL toolchain.

The automotive industry has reached a clear consensus that virtual simulations are crucial to validate the safety of the intended function (SOTIF). This talk presents a conceptual framework that guides these virtual simulations based on ISO 21448. The audience will learn how the industry is addressing the safety of ADAS/AD. Key takeaways will be: • Argumentation framework to ensure SOTIF • Identification of critical scenarios through analysis • Development of validation and verification strategy based on virtual testing • Identification of critical scenarios from real driving data • Standardized interfaces between tools in simulation toolchain

Automated parking has been dominated by ultrasonic-based systems for the last decade. Such systems have fundamental limitation in range and scene understanding, and thus can only cover simple use cases with relatively low success rate. This presentation will introduce a novel approach of fully autonomous parking purely based on camera systems. The system can cope with all scenarios including today-operational design domain and additional challenging use cases such as misalignment, narrow, small/thin obstacles (pole, baby, trolley, etc) and cluttered environments (bushes, nets, fence) with and without park marking. The success rate of the proposed system is much higher than the existing systems.

Oxbotica is advancing self-driving capabilities through the development and implementation of three key products: Oxbotica Driver, Oxbotica Cloud and MetaDriver. This presentation will outline how we are preparing for the safe and successful commercial deployment of these products and will offer insights into our safety management system, preparing for regulatory authorization and ensuring continued safety throughout operations.

In the presentation we provide the basic requirements from ISO 26262 for tool qualification and library qualification. We explain how the tool confidence level is computed and provide examples. We present the arguments for and against tool qualification: Either the tools have to be used "carefully" or they have to be validated. Since library software is part of the product the requirements are higher and will be also presented in the talk, together with reccommendations how to identify only the used functions in order to reduce qualification efforts. We present a set of documents that help to pass every assessment

We present a new paradigm for Autonomous Vehicles (AV). Main approach today consists in cutting situations into hundreds of use-cases and developing a control law for every use-case. We propose a disruption: assessing prudence level of vehicle behavior in its environment, done by an Artificial Intelligence. We explain how this AI works. We show that once prudence level is assessed in real time during driving, one can use it: vehicle behavior must be automated in order to keep prudence at a high level. We show several applications to AV, including STELLANTIS longitudinal automation results with this new AI technology.

Its essential that we can test our ADAS and AV systems on a wide range of scenarios covering nominal, critical and edge cases. This must be done in a deterministic simulation environmen so that the same test can be run in repeatable manner with the only variable being the ADAS/AV controller behaviour. AVSandbox provides a deterministic and sensor realistic simulation environment that allows the ADAS/AV controller to be immersed into the virtual world and tested on a diverse range of scenarios to identify failure modes and focus the development effort to improve safety.

Kontrol has developed a technology and software platform to support requirements processing and system validation. Based on the example of R157, the analysis and processing of a set of legal requirements and corresponding technical requirements is demonstrated. Kontrol enables end-to-end compliance across the value chain. Kontrol illustrates the concept of a digital loop and exemplary virtual homologation of ALKS by integrating Kontrol’s software in a simulator for virtual validation. This helps to further accelerate the deployment of automated vehicle functions and adjacent business models.

To achieve the required reliability and safety levels of ADAS and Autonomous Driving, the industry needs to scale test coverage capacity to execute an ever-expanding range of scenario-based Hardware-in-the-Loop tests utilizing both prerecorded data acquired from test drives and virtual data from high-fidelity environmental simulation software. Testing infinite but real-world scenarios requires switching back and forth between Replay and HIL to maximize test coverage as gaps are not acceptable. Learn how to close a critical gap in direct fault injection into the Mobileye chipset using ECU release software instead of HIL mode, to release safer products to the market faster.

For camera-based automobile Advanced Driver Assistant Systems (ADAS) high image quality levels are essential for autonomous driving levels 3 and higher. This makes improvements in optical quality control of windshields necessary for automotive manufacturers and ADAS developers & glass processors around the world. Established measurement of the pure optical distortions (optical power) in milli diopter (mdpt) are not effective for hyper focal distance. LaVision shows a new test system for quantification of the spatial frequency response (SFR) of the windshield in the ADAS camera area. Impacts of multilayer windshield glasses to image quality are analyzed.

BIM and digital twin are buzz-words in the development and maintenance of public assets, automotive manufacturers invest often in smart city applications and see a merge between autonomous vehicles and public assets maintenance. This paper describes the evolution from digitization for purely automotive development to digitization that support that merge. (Autonomous) vehicles evolve to data source for which use cases extend the pure automotive market. - Automotive use cases for digitizing roads - automotive as data provider for road authorities - drastic change in the landscape of future road surveying

Real-time sensor calibration involves performing calibration tasks on the sensor itself, using local data and computing resources. It improves accuracy, reliability and safety. This is important in applications requiring rapid or real-time responses, like autonomous vehicles, industrial automation, robotics, and more. This presentation will focus on multi-sensor calibration with an emphasis on LiDAR, camera, radar and IMU sensors

Automotive engineers need accurate and reliable vehicle measurements for open-road vehicle testing. Inertial navigation systems provide reference data for position, orientation and vehicle dynamics, but when GNSS signals are blocked or reflected, that data will drift over time. In this presentation, OxTS will present case study data which demonstrates a new breakthrough in open-road INS performance, where LiDAR odometry data and INS navigation data have been combined into one measurement stream, minimising drift in GNSS-denied test areas.

Validation of AVs requires Petabytes of real-world data and the costs for managing the data in the cloud are immense. A key challenge is to access, analyze, and structure the data to automatically identify relevant events. It is important to understand what data has already been recorded, what is redundant and could potentially be disposed of or stored in lower-cost storage options and what data is relevant and still needed to test perception algorithms. We will present the data pipeline and show how sensor data management can help to accelerate the validation and deployment of autonomous vehicles on the road.

AI-based perception alone cannot ensure the safety of automated vehicles. The presentation discusses the safety aspects of typical perception and planning stacks. It proposes a complementary safety channel to address currently used algorithms' weaknesses. This safety channel explicitly handles unsafe and unknown objects and ensures that the automated vehicle never collides with any of them.

It is not acceptable that a safety critical vehicle-to-everything communication device sends the first message right before the accident is about to happen. We need to test these devices thoroughly all along the lifecycle of the vehicle. As we want to rely on vehicles and smart city components not only for transportation, but also as a connected network device, the testing must cover several wireless standards: 4G, 5G, WLAN, Bluetooth, V2X, UWB. This is a wide spectrum to cover, so one must apply innovative test tools to keep costs low and test coverage high.

The key for the success of autonomous driving is guaranteeing safety. Vehicle positioning based on GNSS, hybridized with other sensors like inertial measurement units and wheel odometers, is a critical input to the decision-making algorithm which is used within autonomous driving. Highly accurate positions are not sufficient; in the search of safety, we must ensure that all positioning data produced is free from unexpected errors due to rare hazards no matter how unlikely they are. This is commonly known as the Integrity concept. Integrity is the capability to detect and exclude in real time hazards affecting the positioning data.

With the recent advancement of sensing, AI, and automation technologies, connected automated vehicles have gained significant attention nowadays. As the acceptance of connected automated vehicles increases, innovative connected applications to improve safety and operational efficiency as well as to reduce the possibility of error are also beginning to emerge. We have been developing connected mobility platform FALCON®, which enables improved collaborative applications for connected automated vehicles. This presentation will introduce an innovative routing solution to maximize the time of automated driving for a connected automated vehicle using connected mobility platform FALCON®.

The usage of PNT services/functionalities are increasing. Combined with these, the usage of GNSS correction-systems becomes common and more services providers are coming to the market. Furthermore the services are also used in combination with GNSS based systems for the assessment of the performances of PNT-systems in the scope of type approval e.g. ADAS-UNECE-R152. Thus the need to assess independently the quality, performance and safety of such correction is an increasingly important topic. NavCert conducts certification for this and conducted an internal-project to address topics from ISO-26262. This presentation covers experience from current certifications and topics from the internal project.

The widened application of high level ADAS system towards full automated driving in modern cars together with the rising connectivity demand, such as OTA, lead to increased functional safety (ASIL) and cyber security requirements for all involved components. Providing a precise position with high availability, protection levels, robustness and integrity is a key element for such ADAS systems. This means for a GNSS receiver, as STM's TeseoAPP family, besides robustness against jamming, spoofing, meaconing attacks, to cover additionally safety and security standards as ISO26262 and ISO/SAE 21434, complemented by secure boot capabilities to prevent manipulation of firmware.

Localization is a vehicle’s ability to identify where it is in the world. For autonomous vehicles, accurately and quickly locating themselves in their environment is critical. The autonomous sensor suite consists of many sensors – including but not limited to optical, ranging and inertial – providing relative position. High-accuracy GNSS is the only absolute position sensor to provide the accuracy and confidence that autonomy requires. Sensors are only as good as the confidence in the output. This session will look at how integrating integrity into the autonomous sensor suite improves overall system safety.