Augmenting Human-Robot Collaboration for Real-Time Path Optimization

Abstract

This paper describes a new framework for interaction between humans and robots, focusing on bidirectional data sharing between them in manufacturing environments. The research includes the integration of sensing systems through an Augmented Reality (AR) interface, which functions as a medium for sharing real-time information between humans and robots. One direction of information transmission involves humans communicating inputs such as tasks or interventions to the control systems of robotic mechanisms using hand-gesture and voice-command recognition offered by the AR interface. In the other direction, humans can perceive the robots’ intended motions and actions before execution in three-dimensional holograms created through light diffraction via the AR platform. The new bidirectional interaction integrates human inputs with robots’ control strategies and to enhance human perception of the robotic setting, where humans and robots can better coordinate their operations. Humans use AR to visualize robotic paths and understand their tasks for safety and oversight. They can make decisions by selecting, modifying, or approving robotic actions. Robots, equipped with kinematic, dynamic, and path-planning algorithms, generate path recommendations based on the task requirements, which humans can evaluate and use to make final decisions. The human perceives and evaluates paths calculated by the robot and can select those that meet specific priorities, such as safety or efficiency