Leveraging Stereo-Camera Data for Real-Time Dynamic Obstacle Detection and Tracking

TL;DR

This paper introduces a real-time system for detecting and tracking dynamic obstacles using stereo-camera data, suitable for resource-constrained robots, with high accuracy in indoor and outdoor environments.

Contribution

It presents a novel real-time approach for dynamic obstacle detection and tracking using stereo cameras, optimized for low-power robotic platforms.

Findings

Achieves 85.3% MOTA in dynamic object tracking

Reaches 96.9% precision in static object detection

Operates in real-time on consumer-grade hardware

Abstract

Dynamic obstacle avoidance is one crucial component for compliant navigation in crowded environments. In this paper we present a system for accurate and reliable detection and tracking of dynamic objects using noisy point cloud data generated by stereo cameras. Our solution is real-time capable and specifically designed for the deployment on computationally-constrained unmanned ground vehicles. The proposed approach identifies individual objects in the robot's surroundings and classifies them as either static or dynamic. The dynamic objects are labeled as either a person or a generic dynamic object. We then estimate their velocities to generate a 2D occupancy grid that is suitable for performing obstacle avoidance. We evaluate the system in indoor and outdoor scenarios and achieve real-time performance on a consumer-grade computer. On our test-dataset, we reach a MOTP of $0.07 \pm 0.07m$, and a MOTA of $85.3\%$ for the detection and tracking of dynamic objects. We reach a precision of $96.9\%$ for the detection of static objects.