Systematic and Realistic Testing in Simulation of Control Code for Robots in Collaborative Human-Robot Interactions

摘要: Industries such as flexible manufacturing and home care will be transformed by the presence of robotic assistants. Assurance safety functional soundness for these systems require rigorous verification validation. We propose testing in simulation using Coverage-Driven Verification (CDV) to guide process an automatic systematic way. use a two-tiered test generation approach, where abstract sequences are computed first then concretized (e.g., data variables instantiated), reduce complexity problem. To demonstrate effectiveness our we developed testbench code, running ROS-Gazebo, that implements object handover part human-robot interaction (HRI) task. Tests generated stimulate robot’s code realistic manner, through stimulating human, environment, sensors, actuators simulation. compare merits unconstrained, constrained model-based achieving thorough exploration under test, interesting combinations interactions. Our results show CDV combined with achieves very high degree automation simulation-based control robots HRI.