Infrastructure Enabled Autonomy Acting as an Intelligent Transportation System for Autonomous Cars
MetadataShow full item record
Autonomous cars have the ability to increase safety, efficiency, and speed of travel. Yet many see a point at which stand-alone autonomous agents populate an area too densely, creating increased risk - particularly when each agent is operating and making decisions on its own and in its own self-interest. The problem at hand then becomes how to best implement and scale this new technology and structure in such a way that it can keep pace with a rapidly changing world, benefitting not just individuals, but societies. This research approaches the challenge by developing an intelligent transportation system that relies on an infrastructure. The solution lies in the removal of sensing and high computational tasks from the vehicles, allowing static ground stations with multi sensor-sensing packs to sense the surrounding environment and direct the vehicles safely from start to goal. On a high level, the Infrastructure Enabled Autonomy system (IEA) uses less hardware, bandwidth, energy, and money to maintain a controlled environment for a vehicle to operate when in highly congested environments. Through the development of background detection algorithms, this research has shown the advantage of static MSSPs analyzing the same environment over time, and carrying an increased reliability from fewer unknowns about the area of interest. It was determined through testing that wireless commands can sufficiently operate a vehicle in a limited agent environment, and do not bottleneck the system. The horizontal trial outcome illustrated that a switching MSSP state of the IEA system showed similar loop time, but a greatly increased standard deviation. However, after performing a t-test with a 95 percent confidence interval, the static and switching MSSP state trials were not significantly different. The final testing quantified the cross track error. For a straight path, the vehicle being controlled by the IEA system had a cross track error less than 12 centimeters, meaning between the controller, network lag, and pixel error, the system was robust enough to generate stable control of the vehicle with minimal error.
SubjectInfrastructure Enabled Autonomy
Intelligent Transportation System
Burch, Austin Jess (2018). Infrastructure Enabled Autonomy Acting as an Intelligent Transportation System for Autonomous Cars. Master's thesis, Texas A & M University. Available electronically from