Object Servoing of Differential-Drive Robots

TL;DR

This paper introduces a new object servoing scheme for differential-drive robots that estimates relative pose using feature points, enabling accurate parking near movable objects with low computational cost.

Contribution

A novel object servoing scheme that estimates relative pose online and guides differential-drive robots to park accurately near movable objects.

Findings

Effective in asymptotic parking at predefined relative poses

Low online computational cost

Validated through experimental results

Abstract

Due to possibly changing pose of a movable object and nonholonomic constraint of a differential-drive robot, it is challenging to design an object servoing scheme for the differential-drive robot to asymptotically park at a predefined relative pose to the movable object. In this paper, a novel object servoing scheme is designed for the differential-drive robots. Each on-line relative pose is first estimated by using feature points of the moveable object and it serves as the input of an object servoing friendly parking controller. The linear velocity and angular velocity are then determined by the parking controller. Experimental results validate the performance of the proposed object servoing scheme. Due to its low on-line computational cost, the proposed scheme can be applied for last mile delivery of differential-drive robots to movable objects.