Pseudospectral Optimal Control: A Clear Road for Autonomous Intelligent Path Planning
Kevin Bollino, L. Ryan Lewis, Pooya Sekhavat, I. Michael Ross · AIAA Infotech@Aerospace 2007 Conference and Exhibit · 2007
This paper captures the essence of intelligent path planning by posing the problem in a framework based on optimal control theory. Design requirements for autonomous vehicles call for on-board intelligence capable of making timely decisions, performing tasks in a “smarter ” fashion, and ultimately accomplishing missions with extreme accuracy. This is the definition of an optimal control problem! Indeed, most, if not all, motion planning problems can be formulated and solved using optimal control techniques. Motivated by the significant advancements in optimal control techniques over the last decade, we demonstrate the broad range of unmanned systems that can operate both optimally and autonomously by solving path-planning problems using pseudospectral methods. I.