Clutter Resilient Occlusion Avoidance for Tightly-Coupled Motion-Assisted Detection

TL;DR

This paper introduces a clutter resilient occlusion avoidance method for motion-assisted detection that actively plans robot viewpoints to reduce occlusion in cluttered environments, improving detection robustness.

Contribution

The paper presents a novel CROA framework that enhances MAD by minimizing occlusion probability using convex-concave procedures and bilevel optimization, specifically designed for cluttered scenarios.

Findings

CROA outperforms existing MAD schemes in point density and detection accuracy.

The method is robust in cluttered urban driving environments.

Experimental results validate improved detection metrics under various conditions.

Abstract

Occlusion is a key factor leading to detection failures. This paper proposes a motion-assisted detection (MAD) method that actively plans an executable path, for the robot to observe the target at a new viewpoint with potentially reduced occlusion. In contrast to existing MAD approaches that may fail in cluttered environments, the proposed framework is robust in such scenarios, therefore termed clutter resilient occlusion avoidance (CROA). The crux to CROA is to minimize the occlusion probability under polyhedron-based collision avoidance constraints via the convex-concave procedure and duality-based bilevel optimization. The system implementation supports lidar-based MAD with intertwined execution of learning-based detection and optimization-based planning. Experiments show that CROA outperforms various MAD schemes under a sparse convolutional neural network detector, in terms of point density, occlusion ratio, and detection error, in a multi-lane urban driving scenario.