College - Author 1
College of Engineering
Department - Author 1
Electrical Engineering Department
Degree Name - Author 1
BS in Electrical Engineering
College - Author 2
College of Engineering
Department - Author 2
Electrical Engineering Department
Degree - Author 2
BS in Electrical Engineering
College - Author 3
College of Engineering
Department - Author 3
Electrical Engineering Department
Degree - Author 3
BS in Electrical Engineering
College - Author 4
College of Engineering
Department - Author 4
Electrical Engineering Department
Degree - Author 4
BS in Electrical Engineering
Date
6-2023
Primary Advisor
Xiao-Hua Yu, College of Engineering, Electrical Engineering Department
Abstract/Summary
As the world and industry look to automate and optimize processes, solutions such as robotics, artificial intelligence, and algorithms emerge to fulfill these needs. Our project builds on a previous senior project, “Robot Path Planning System – Berry et al. (2021)”, which implements an Artificial Potential Field (APF) path-finding algorithm with higher order Euclidean functions and attempts to successfully utilize a neural network-based reinforcement-learning algorithm.
The A* algorithm is a flagship path finding algorithm to determine the shortest distance from a start and end point in a two-dimensional environment. The main issue presented with this path finding algorithm is the time-consuming computational cost of determining and executing the shortest path. The APF algorithm uses a potential field map to determine the path of from the start to end point. Although the APF algorithm is more susceptible to incomplete path determinations from using local determinations rather than global determinations in A* path finding, the computational cost is significantly less. This poses an immense potential for APF utilization in an optimized state.
One improvement posed to the APF algorithm is the virtual obstacle method. This allows the robot to overcome certain obstacles where the algorithm would stall in a fixed position. With the virtual obstacle method, the robot can regain movement when in a fixed position from the local environment. The team implements, optimizes, and tests the basic APF algorithm and improved APF algorithm with the virtual obstacles’ method for path planning. Evaluation is conducted on the APF algorithm including performance testing under unique static environments. Each of these environment simulation applications pushes the APF to function beyond the initial limitations with the virtual obstacle’s method, passing new cases and old limitations.
Our deliverables make expansions and improvements to the capabilities of the APF algorithm, as well as compare our results for a thorough environmental analysis with the A* path finding algorithm. Our results, (with runtime, computational iterations, and test success) display that the APF and A* algorithms to have situational advantages and disadvantages for the needs of the user and demands and scale of the environment.
URL: https://digitalcommons.calpoly.edu/eesp/608