LAUSR

Abstract Service robots meet numerous tasks-level commands with uncertain relationship between targets and obstacles, which introduces plenty of programming work involved with obstacle avoidance in robots’ operations for non-expert users. Hence an obstacle avoidance algorithm based on task semantic object matrix(SOM) was introduced to tell targets from obstacles in this paper. With this strategy, petri net was applied to decompose the robots’ complex operation tasks into sequential subtasks, which can be practical for finite state machine. And the SOM related to sequential subtasks was generated synchronously. Based on the SOM, the point cloud segmentation, collision geometry modeling and the transformation between targets and obstacles can be conducted for robots’ operations. Experimental results indicate that the SOM can be updated within 0.25s. And the experiments demonstrated that this algorithm is feasible to help non-expert users conduct robots’ task-based operations.