SURF 2018: Experimental verification of a semi-autonomous vehicle design based on human intention
2018 SURF: project description
- Mentor: Richard M. Murray
- Co-mentor: Jin Ge
While self-driving features such as adaptive cruise control and lane-keeping systems have been implemented in production vehicles, there are still many obstacles for the implementation of fully automated vehicles in real traffic. However, semi-autonomous driving can be achieved with few legal or policy obstacles. In particular, we are interested in a semi-autonomous vehicle design that can detect human intentions and interact safely with other road users. Such a vehicle design not only can enhance traffic safety, but it may also significantly benefit people who are physically challenged to drive.
In this study, we may first use data-driven methods to detect the intention of the human operator on-board and also the intention of other road users nearby. Then a higher-level advisory controller can be designed to execute, modify, or override the human intention, similar to the flight envelope protection system in aviation. Combining the intention detection, advisory controller and the lower-level driving/steering control, we have a semi-autonomous vehicle design.
The semi-autonomous vehicle design will be implemented and tested on F1/10 platform. The testing scenarios may include multi-lane driving, parking, or navigating intersections where other human-operated vehicles are around. For the controller design, general knowledge about planar rigid-body dynamics and PID control are desired. For the experimental implementation, programming experience with Python and ROS is needed.
 S. A. Seshia, D. Sadigh and S. S. Sastry, "Formal methods for semi-autonomous driving," 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC), San Francisco, CA, 2015, pp. 1-5.
 B. Okumura et al., "Challenges in Perception and Decision Making for Intelligent Automotive Vehicles: A Case Study," in IEEE Transactions on Intelligent Vehicles, vol. 1, no. 1, pp. 20-32, March 2016.
 Duo Han, Yilin Mo and Richard M. Murray, “Synthesis of Distributed Longitudinal Control Protocols for a Platoon of Autonomous Vehicles”, Technical report done at Caltech, 2014.
 J. I. Ge, G. Orosz, and R. M. Murray. Connected cruise control design using probabilistic model checking. Proceedings of the American Control Conference, 4964-4970, IEEE, 2017.
 F110 autonomous race cars. <http://mlab-upenn.github.io/f110/index.html>