Distilling Privileged Information for Dubins Traveling Salesman Problems with Neighborhoods

TL;DR

This paper introduces a two-phase learning approach for Dubins TSP with Neighborhoods that significantly accelerates solution generation and outperforms existing methods by leveraging privileged information and expert demonstrations.

Contribution

It proposes a novel two-phase learning framework combining reinforcement and supervised learning to efficiently solve DTSPN problems using privileged information.

Findings

Solution is about 50 times faster than LKH heuristic

Outperforms other imitation learning and RL methods

Effectively senses all task points in complex scenarios

Abstract

This paper presents a novel learning approach for Dubins Traveling Salesman Problems(DTSP) with Neighborhood (DTSPN) to quickly produce a tour of a non-holonomic vehicle passing through neighborhoods of given task points. The method involves two learning phases: initially, a model-free reinforcement learning approach leverages privileged information to distill knowledge from expert trajectories generated by the LinKernighan heuristic (LKH) algorithm. Subsequently, a supervised learning phase trains an adaptation network to solve problems independently of privileged information. Before the first learning phase, a parameter initialization technique using the demonstration data was also devised to enhance training efficiency. The proposed learning method produces a solution about 50 times faster than LKH and substantially outperforms other imitation learning and RL with demonstration schemes, most of which fail to sense all the task points.