Optimal Path Planning of a Solar-Powered Unmanned Ground Vehicle in an Unknown Solar Environment with Multi-Objective Optimization

• Main Authors: Strebe, Luke, Lee, Kooktae
• Format: Conference Proceeding
• Language: English
• Subjects: Batteries; Energy Maximization; Environmental monitoring; Land vehicles; Multi-Objective Optimization; Optimal Path Planning; Optimization; Sensors; Simulation; Solar Power; Trajectory; Unmanned Ground Vehicle
• ISSN: 2378-5861
• DOI: 10.23919/ACC60939.2024.10644455

Solar Powered Unmanned Ground Vehicles (SPUGV) can be used for long-term environmental monitoring of an area, however, there is limited research regarding battery maximizing path planning in an unknown solar environment. In this paper, a novel approach for optimal path planning in a completely unknown solar environment is investigated using Multi-Objective Optimization (MOO). The Feasibility Space Path Planner (FSPP) is proposed to update the path of the SPUGV in a receding-horizon fashion as it gains information about solar availability in the environment from onboard sensors. The use of MOO in path planning provides optimal path planning for maximizing the final battery of the SPUGV. The path that provides the maximum battery while also minimizing the distance towards the goal is chosen, and re-evaluated at each time step the SPUGV moves. Therefore, the SPUGV will create battery maximization optimal trajectories without any prior information about an area. Simulation results concluded drastically improved final battery values using the FSPP algorithm compared to straight path trajectories toward a goal.