Greedy Clustering-Based Algorithm for Improving Multi-point Robotic Manipulation Sequencing

TL;DR

This paper introduces a greedy clustering-based algorithm that improves multi-point robotic manipulation sequencing by balancing solution quality and computational efficiency, validated through simulations and real robot experiments.

Contribution

The paper presents a novel greedy clustering algorithm that enhances sequencing efficiency for robotic manipulation, addressing real-time constraints in multi-point tasks.

Findings

Significant reduction in planning time compared to baseline algorithms

Validated improvements on a UR5 robot in real-world experiments

Effective balance between solution quality and computational speed

Abstract

The problem of optimizing a sequence of tasks for a robot, also known as multi-point manufacturing, is a well-studied problem. Many of these solutions use a variant of the Traveling Salesman Problem (TSP) and seek to find the minimum distance or time solution. Optimal solution methods struggle to run in real-time and scale for larger problems. In online planning applications where the tasks being executed are fast, the computational time to optimize the ordering can dominate the total execution time. The optimal solution in this application is defined as the computational time for planning plus the execution time. Therefore, the algorithm presented here balances the quality of the solution with the total execution time by finding a locally optimal sequence. The algorithm is comprised of waypoint generation, spatial clustering, and waypoint optimization. Significant improvements in time reduction were seen and validated against a base case algorithm in simulation and on a real UR5 robot.