Research on Human-Robot Cooperative Target Recognition for Spatial Sampling Task

Objective—A new type human-robot collaborative target recognition mode was designed to improve the performance of spatial sampling tasks. Methods—The new type not only presented the target attribute information recognized by the robot sensors to humans, but also expressed the credibility of the information determined by the algorithm through a designed interface. An experimental platform for simulating the physical and task environment of the lunar surface were constructed, on which the new type collaborative recognition mode as well as a traditional type mode were designed and simulated respectively. Results—The experiment results showed that the new type human-robot collaborative target recognition mode achieved a better accuracy of target recognition. Conclusions—In the new mode constructed in this research, the robot processed and presented the attribute information in a humanized manner which matched the human cognitive architecture, thereby achieving a better human-robot collaborative performance.