Bridging the Gap between Open Source Software and Vehicle Hardware for Autonomous Driving

Abstract

Although many research vehicle platforms for autonomous driving have been built in the past, hardware design, source code and lessons learned have not been made available for the next generation of demonstrators. This raises the efforts for the research community to contribute results based on real-world evaluations as engineering knowledge of building and maintaining a research vehicle is lost. In this paper, we deliver an analysis of our approach to transferring an open source driving stack to a research vehicle.We put the hardware and software setup in context to other demonstrators and explain the criteria that led to our chosen hardware and software design. Specifically, we discuss the mapping of the Apollo driving stack to the system layout of our research vehicle, fortuna, including communication with the actuators by a controller running on a real-time hardware platform and the integration of the sensor setup. With our collection of the lessons learned, we encourage a faster setup of such systems by other research groups in the future.

Optimal Behavior Planning for Autonomous Driving: A Generic Mixed-Integer Formulation

“…This property is certainly a favorable feature for validation and certification. This will enable us in the future to apply the MIQP behavior planning model in real-road driving scenarios using our institute's research vehicle [14]. Future work will expand this framework to the multi-agent case for cooperative planning as well as to investigate the benefits from logical constraints for the formulation of traffic rules.…”

“…Collecting all constraints from above, the final optimization problem can be written as minimize (10) subject to (11), (12), (13), (14), (15), (16), The formulation is a standard MIQP model that can be solved with an off-the-shelf solver. For a faster optimization solution, we bound acceleration a 9 and jerk u 9 with constant values a, a and u, u respectively.…”

Linear Differential Games for Cooperative Behavior Planning of Autonomous Vehicles Using Mixed-Integer Programming

“…In combination with our previous work [12], which proved correctness in the model of the vehicle kinematics and the obstacle avoidance, we have successfully demonstrated the feasibility in simulation. As a next step, we plan to apply the formulation in realroad driving scenarios using the institute's research vehicle [17]. For this to be feasible regarding computation time, we need to find a trade-off which agents to denote as interacting and which others to predict maneuvers for solely.…”

Cooperative Multi-Vehicle Behavior Coordination for Autonomous Driving

“…As the next steps, we plan to further emphasize the interaction-awareness to other traffic participants by improving the simple implementation of the behavior reflection step by evaluating game theoretic or learning-based approaches. To bring the method to our institute's autonomous driving prototype vehicle [17], we further aim for smoother motion profiles and a more efficient implementation using e.g. parallelization or a tailored optimization solver.…”