A small-scaled testbed for integrated manual and autonomous driving

The purpose of this thesis is to create a small-scaled testbed for integrated manual and autonomous driving. The research task can be divided into two main parts. The first part is the development of an Intelligent Transportation System (ITS) testbed which consists of an arena; an indoor localization system; Radio Controlled (RC) cars; a steering wheel system and roadside monitoring facilities. All these hardware are managed by a server. This server controls the RC cars, collects data from the indoor localization system, the motion sensor on the RC car and the steering wheel system, and streams out the data so that a remote client in other place in the world can access and process the data instantaneously. The second part discusses the mechanism of switching between manual and autonomous driving based on a drowsiness detection algorithm. A user would manually drive the RC car. When he is sleepy or drowsy, the RC car would be autonomously controlled by the server. If the driver can maintain his consciousness, the server would let him keep driving the RC car manually. The drowsiness detection is performed by a client application which would send the results back to the server so that the server can have appropriate control to the RC car. The experiments were conducted on the ITS testbed to demonstrate the proposed manual/autonomous driving mechanism.