Provably safe design of driver-assist systems through hybrid automata with hidden modes

In this thesis, I consider the problem of collision avoidance between two vehicles approaching an intersection. These vehicles are human driven and one or both are equipped with an on-board driver assist system that provides warnings and can apply automatic braking/throttle when needed. This type of system will establish an intermediary step in the progression towards fully autonomous vehicles. It will allow human drivers to retain control of their vehicles while providing the guidance for drivers to apply the necessary inputs to prevent collisions before autonomous control becomes necessary. A formal approach to the design of the driver assist system is taken, employing a hybrid automaton model. This model has hidden modes, which arise from the driver making decisions about whether or not to follow the provided warnings. As a consequence, the driver assist system design is formulated as a safety control problem for a hybrid automaton with hidden modes. The solution approach is based on a mode estimator that keeps track of the possible driver decisions and, on their basis, provides warning and control inputs that ensure safety. The resulting algorithm is computationally efficient as it leverages the order preserving properties of the vehicle dynamics.

Department dc:contributor.department

Massachusetts Institute of Technology. Dept. of Mechanical Engineering.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2012

Author dc:creator

• Palas, Cassidy Martin

Advisor dc:contributor.advisor

• Domitilla Del Vecchio.

• Mechanical Engineering.

Statement dc:rights

• M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/74935

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/74935