Safety in human-robot collaborative manufacturing environments: Metrics and control

TLDR

Industrial robotics now allow humans and robots to share workspaces, but conventional motion generation algorithms may be insufficient for such collaborative settings. The study proposes a kinematic control strategy that enforces safety while maximizing robot productivity. An optimization‑based real‑time algorithm generates the robot’s motion, treating safety as a hard constraint, and the approach is validated on a dual‑arm 7‑DOF robot performing a manipulation task. The experimental validation on a dual‑arm 7‑DOF robot performing a manipulation task demonstrates the feasibility of the approach.

Abstract

New paradigms in industrial robotics no longer require physical separation between robotic manipulators and humans. Moreover, in order to optimize production, humans and robots are expected to collaborate to some extent. In this scenario, involving a shared environment between humans and robots, common motion generation algorithms might turn out to be inadequate for this purpose. This paper proposes a kinematic control strategy which enforces safety, while maintaining the maximum level of productivity of the robot. The resulting motion of the (possibly redundant) robot is obtained as an output of an optimization-based real-time algorithm in which safety is regarded as a hard constraint to be satisfied. The methodology is experimentally validated on a dual-arm concept robot with 7-DOF per arm performing a manipulation task.