A Framework for Collision-Tolerant Optimal Trajectory Planning of Autonomous Vehicles

TL;DR

This paper introduces a collision-tolerant trajectory planning framework for autonomous vehicles, using mixed integer programming to optimize paths that include planned collisions, thereby improving performance over traditional collision-free methods.

Contribution

It presents a novel MIP-based optimization approach that incorporates damage quantification functions to enable collision-tolerant trajectory planning for autonomous vehicles.

Findings

Collision-tolerant trajectories outperform collision-free ones in simulations

Damage functions influence trajectory optimization effectively

Mixed integer programming is practical for collision-inclusive path planning

Abstract

Collision-tolerant trajectory planning is the consideration that collisions, if they are planned appropriately, enable more effective path planning for robots capable of handling them. A mixed integer programming (MIP) optimization formulation demonstrates the computational practicality of optimizing trajectories that comprise planned collisions. A damage quantification function is proposed, and the influence of damage functions constraints on the trajectory are studied in simulation. Using a simple example, an increase in performance is achieved under this schema as compared to collision-free optimal trajectories.