Internship Model Based/Data Driven Control Strategies For Vehicle Motion Control

Your team and your responsibilities Team/division descriptionThe internship will be developed in TME R&D, Chassis group. Chassis engineering division is responsible for the:Development of Chassis systems and components (Brake, Steering, Suspension, Tyre, Shift, Pedals, …) for local production, utilising European technologies and suppliers to achieve the highest driving performance and quality standards at competitive cost.Vehicle dynamic performance development through simulation and testing of systems and vehicles.Investigation and Predevelopment of new European technology for automotie chassis systems and components. Research and application of state-of-the-art methodologies for chassis systems and chassis electronic systems’ design and validation, enabling most competitive performance and efficient development processes while securing our quality and safety standards.We are inspired by the Toyota Way precepts, and we challenge to provide final customers with best quality products with the continuous improvement mindset.Your projectThe internship will deal with the development of advanced and novel control strategies for the lateral and longitudinal control of a generic AD application, focusing on the safety and the performance maximization on the no-linear domain of motion (. evasive / emergency manoeuvres) The intern will take care to contribute to an existing line of investigation, based on a predeveloped MPC framework in a simulation environment that controls the brake and steering of a vehicle based on a given path plan. The goals of the intern will deal with Increasing robustness of the vehicle motion vs the noises in real world (sensors, actuators, road geometry, wind…) and with the change of vehicle characteristics vs the ideal MPC model through a continuous supervision of the model characteristics (. Learning based MPC).Developing a practical solution which considers the practical constraints of implement-ability in an AD application, such as data quality, state estimation accuracy, real time limitations and efficiency for development and tuning processes.That tasks of the intern will include, but not limited to:study of prior existing MPC development (formulation, simulation platform)study of prior implemented data driven control robustness measures: Learning Based MPC frame, Reference Governor re-plannerfurther review on alternative robustness measures (through literature review)complete the definition of the range for robustness concerns in current MPC frame (incl. upstream system deviations from ideal world, such as sensor physical modelling, path planning inaccuracies and errors in estimators)propose and implement applications to mitigate the MPC limitationsdecide solution to the motion control robustness problem. Your Profile You are fluent in English (TME’s business language)You are student in last year of a Masters Degree in Mechatronics, Control Systems or related domain.You have interest in Control Engineering, Vehicle Dynamics and Autonomous Vehicle systems.You have strong knowledge in Control Theory and Robust Control techniques as well in Dynamic and Vehicle Dynamics simulations and related tools (Matlab/Simulink, CarMaker, SimPack, Acado) You have strong coding skills (Python, Matlab) and highly knowledgeable in Data Science and Machine Learning.You have good communication skills and can work in autonomous way within a multicultural teamStarting date AsapDuration 6 monthsConfidentialityDue to business requirement, not all performed projects can be reflected in the internship report. This issue needs to be discussed with candidate/school in advance.