Motion planning for persistent traveling solar-powered unmanned ground vehicles
Adam Kaplan, Nathaniel Kingry, Justin Van Den Top, Kishan Patel, Ran Dai, David Grymin
Abstract
This paper examines a mission planning problem for a solar-powered unmanned ground vehicle (UGV) which requires the vehicle to visit a series of objective points in minimal time subject to a strict net-energy change constraint. Though related to the Traveling Salesperson Problem, the mission planning problem discussed herein imposes further complexity through additional coupled mixed-variable sets and the strict energy constraint. A scalar field representing the solar radiation of the mission environment is first characterized from a visual-spectrum image. A cascaded particle swarm optimization algorithm, coupled with the integer linear programming technique, is used to generate a time-optimized motion plan and power schedules for the UGV, which guides it to visit the assigned objective points with optimized sequence and paths, and then return to its starting location and orientation while guaranteeing compliance with the net energy gain constraint.
BibTeX
@inproceedings{iros2016_motionplanningfo,
title = {Motion planning for persistent traveling solar-powered unmanned ground vehicles},
author = {Adam Kaplan and Nathaniel Kingry and Justin Van Den Top and Kishan Patel and Ran Dai and David Grymin},
booktitle = {IROS 2016},
year = {2016}
}