M.S. (Master of Science)
Department of Electrical Engineering
Autonomous vehicles are gradually entering our daily lives. The goal of fully autonomous commercially available vehicles is becoming closer to reality each day as the contributions from researchers and various institutions are being added to the overall body of knowledge. Object detection is a critical component of an autonomous or semi-autonomous vehicle and draws extensively on results from many fields such as image processing and statistics. In this thesis, we consider ideas from the study of real-time computing and control systems to present a novel method of real-time adaptive object detection. We present a conceptual framework of the method as it applies to an automated vehicle control system. The application controls an object recognition detection sequence through using the aggregate channel features (ACF) detection algorithm. Our proposed method incorporates awareness of computational resources and feedback from the vehicle motion planner as inputs to the perception algorithm. We provide a complete model for analysis and simulation in MATLAB and Simulink environment. Experimental results are provided across a set of parameters, showing results consistent with the expectations in the proposed framework. The results show promising performance in the simulated scenario of highway driving on a straight road. Several possibilities for extension of the model are possible.
Wolfe, Christopher, "Adaptive Object Detection for Autonomous Vehicles" (2020). Graduate Research Theses & Dissertations. 7787.
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.