Thomas Pettersen

Augmented reality for programming industrial robots

Existing practice for programming robots involves teaching it a sequence of waypoints in addition to process-related events, which defines the complete robot path. The programming process is time consuming, error prone and, in most cases, requires several iterations before the program quality is acceptable. By introducing augmented reality technologies in this programming process, the operator gets instant real-time, visual feedback of a simulated process in relation to the real object, resulting in reduced programming time and increased quality of the resulting robot program. This paper presents a demonstrator of a standalone augmented reality pilot system allowing an operator to program robot waypoints and process specific events related to paint applications. During the programming sequence, the system presents visual feedback of the paint result for the operator, allowing him to inspect the process result before the robot has performed the actual task.