M.S. (Master of Science)
Department of Electrical Engineering
Autonomous vehicles have been the subject of intense research leading to great improvements in safety, comfort, accessibility, efficiency, etc. Path planning which provides the autonomous vehicle with a smooth, collision-free, and safe path towards its destination, is a crucial part of autonomous vehicles (AVs). Various path planning techniques have been developed in the past decades that can successfully plan a smooth and safe path in various driving scenarios. Nevertheless, these algorithms perform poorly in critical situations where the vehicle needs to quickly make a decision. To address this issue, a couple of algorithms have been proposed to replace the AV’s conventional path planning algorithm in situations with high a risk of collision as none of the current algorithms are efficient enough to perform the complete path planning for the vehicle. This study proposes a novel risk-aware planning method based on trajectory generation in the Frenet framework which is capable of planning a smooth, safe, and collision-free path for the AV in normal and critical driving situations. This algorithm works based on generating and evaluating a set of potential trajectories using a proposed cost function based on the lateral deviation, time, speed limit, and risk of trajectories. This is determined using a novel risk assessment method based on Fuzzy Logic Algorithms. Achieving a reliable path planning method that performs efficiently in complex environments such as highways, is a key factor. Therefore, a prediction algorithm has been developed in this study to estimate the position, velocity, and trajectory of obstacles in the surrounding environment of the ego vehicle. This thesis concludes that the proposed risk-aware path planning method builds safe, smooth, and efficient trajectories that follow the desired path, estimate the position and velocity of surrounding vehicles, and avoid collisions in various normal or high-risk scenarios.
Gaeini, Shaghayegh, "Risk-Aware Path Planning For Autonomous Vehicles" (2022). Graduate Research Theses & Dissertations. 7052.
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.