Analysis of task-based gestures in human-robot interaction

New developments, innovations, and advancements in robotic technology are paving the way for intelligent robots to enable, support, and enhance the capabilities of human workers in manufacturing environments. We envision future industrial robot assistants that support workers in their tasks, advancing manufacturing quality and processes and increasing productivity. However, this requires new channels of fine-grained, fast and reliable communication. In this research we examined the communication required for human-robot collaboration in a vehicle door assembly scenario. We identified potential communicative gestures applicable to this scenario, implemented these gestures on a Barrett WAM
^TM1
manipulator, and evaluated them in terms of human recognition rate and response time in a real-time interaction. Response time analysis reveals insights into the communicative structure of robot motions; namely, key short gesture segments include the bulk of the communicative information. These results will help us design more efficient and fluid task flow in human-robot interaction scenarios.