RRT Trajectory Planning Approach For Automated Semi-trailer truck Parking

Abstract

The automation of processes, especially in the automotive sector, has experienced an increase in safety and efficiency, and at the same time, it reduces execution times and human errors. Nowadays, conventional logistics has experimented with significant change. Those go from individual transport services using personal vehicles, passing through internal logistics of the warehouse, and finishing with the delivery of goods in different last mile transport and circumstances with mobile robots (e.g. Amazon logistic centers). However, the automation solutions for heavy-duty vehicles during loading and unloading processes are still a research challenge. This work presents a trajectory planning which targets automated semi-trailer trucks. The algorithm considers turning under difficult and tight circumstances, as well as reversing maneuvers. It is based on RRT (Rapidly exploring Random Trees) planning. The validation was done in a simulation environment considering a logistic warehouse. The solution generates the best possible path in a short time (under 500 milliseconds), and it is a promising result for a near-future real-deployment.