An Image-Based Human-Robot Collision Avoidance Scheme: A Proof of Concept

OCCUPATIONAL APPLICATIONS

In modern industrial plants, collisions between humans and robots pose a significant risk to occupational safety. To address this concern, we sought to devise a reliable system for human-robot collision avoidance system employing computer vision. This system enables the proactive prevention of dangerous collisions between humans and robots. In contrast to previous approaches, we used a standard RGB camera, making implementation more convenient and cost-effective. Furthermore, the proposed method greatly extends the effective detection range compared to previous studies, thereby enhancing its utility for monitoring large-scale workplaces.

TECHNICAL ABSTRACT

Background: Human-robot collaboration is becoming increasingly prevalent in the manufacturing industry, but the risk of collisions between human workers and collaborative robots poses safety concerns.

Purpose: We aimed to address these safety concerns and enhance occupational safety by developing a camera-based motion tracking method combined with a collision-avoidance scheme.

Methods: The approach involves a single RGB camera and computer-vision algorithms to track the position of a human worker relative to a collaborative robot. A comprehensive collision-avoidance scheme is then employed, using the worker’s position and motion data for robot action planning and execution. The collision avoidance actions include generating visual and auditory alarm signals and retracting robot arms.

Results: Our preliminary validation demonstrated that the proposed collision avoidance scheme effectively enables a collaborative robot to respond to a human worker approaching from different directions.

Conclusion: The proof-of-concept human-robot collision avoidance scheme presented in this study significantly reduces the risk of collisions and holds potential for improving workplace safety in future human-robot collaboration applications.

Keywords:

• Collaborative robot
• collision avoidance
• computer vision
• robot kinematics
• risk assessment

Conflict of interest

The authors declare no conflict of interest.

Additional information

Funding

This manuscript is based upon work supported by the National Science Foundation under Grant # 2024688.