1

Speaker Details

Speaker Company

Lee Gonzales-Fuentes

Lee is an electronics engineer specializing in telecommunications with a multidisciplinary profile in the fields of RF measurements, microwave engineering and signal processing with a focus on hardware design and applied mathematics (statistics). Currently, she is a staff application engineer at the automotive microcontroller department at Infineon Technologies, working primarily in radar signal processing for advanced driver-assistance systems (ADAS) for 60GHz and 77GHz applications. She is also a reviewer for the IEEE Transactions in Instrumentation and Measurement.

Presentation

Vehicle trajectory tracking using neural network in autonomous driving systems

Trajectory planning and tracking ensure that a car follows the expected route while motion errors are corrected to minimize deviation from the centerline. Diverse model predictive control (MPC) methods are normally used for this purpose. However, the non-linear dynamics of the car can degrade the performance of these methods. Therefore, neural networks (NN) are employed to enhance the MPC considering the linear and non-linear dynamics of the car so that the effects of disturbances are reduced. The traditional and NN enhanced MPC can be implemented in the new AURIXTM TC4xx microcontroller. A comparative assessment of the proposed method with standard methods is also discussed.