A lightweight method for detecting dynamic target occlusions by the robot body

TL;DR

This paper introduces a lightweight, real-time method using virtual depth images from URDF models to detect and handle dynamic target occlusions caused by the robot itself, enhancing perception robustness.

Contribution

The novel approach leverages URDF-based virtual depth images for real-time occlusion detection, improving robot perception during dynamic tasks.

Findings

Effective occlusion detection during robot movement

Improved accuracy in pose estimation and human tracking

Real-time performance demonstrated with a 6-DoF robot

Abstract

Robot vision is greatly affected by occlusions, which poses challenges to autonomous systems. The robot itself may hide targets of interest from the camera, while it moves within the field of view, leading to failures in task execution. For example, if a target of interest is partially occluded by the robot, detecting and grasping it correctly, becomes very challenging. To solve this problem, we propose a computationally lightweight method to determine the areas that the robot occludes. For this purpose, we use the Unified Robot Description Format (URDF) to generate a virtual depth image of the 3D robot model. Using the virtual depth image, we can effectively determine the partially occluded areas to improve the robustness of the information given by the perception system. Due to the real-time capabilities of the method, it can successfully detect occlusions of moving targets by the moving robot. We validate the effectiveness of the method in an experimental setup using a 6-DoF robot arm and an RGB-D camera by detecting and handling occlusions for two tasks: Pose estimation of a moving object for pickup and human tracking for robot handover. The code is available in \url{https://github.com/auth-arl/virtual\_depth\_image}.