Effects of dynamic planning on the adaptability of a cobotic system to human constraints for a cooperative HRI assembly task

The increasing presence of robots in industrial settings has led to a growing interest in human-robot collaboration (HRC). The focus of this report is on the problem of autonomous and automatic task planning for adapting a cobotic system’s plan of actions to specific human factors and constraints. To favor ergonomic choices and reduce muscoloscheletal disorders, the human should have the freedom of choice of his next action. To increase security and reduce mental load, the robot needs to be aware of human constraints and plan also according to them. For this the robot needs to dynamically plan its next actions taking into consideration both factors: human choices and human limits. We propose a framework based on the use of AI planning and PDDL for the dynamic update of the cobot’s plan, in the context of a human robot collaboration assembly task. An experiment is conducted to validate the proposed implementations and assess their usefulness. The experiment involves a collaborative task between humans and robots, evaluating performance, risk handling, and acceptability. The expected results aim to demonstrate the adaptability of the cobotic system, improved human experience, and effective planning.