Sensing-System Reconfiguration: A Comparison of On-Line Methods

This paper investigates the performance of two dispatching approaches applied to the real-time coordination of multiple, mobile sensors. The sensing system is targeted towards the surveillance of objects in the context of autonomous manufacturing systems. Sensors are assigned and manoeuvred to collect data at specific points on the object trajectory. A technique based on reinforcement learning (RL) is compared to a heuristic dispatching method and a system that does not use dispatching at all. Through a number of simulation examples, it is shown that, on average, the RL-based dispatcher achieves very similar, if not slightly better, performance than the heuristic dispatcher. Both approaches appear to provide a benefit over non-dispatching systems, thereby validating the efficacy of the dispatching approach, despite very different underlying implementations. The use of industrial robots in an automated manufacturing environment typically requires presentation of the objects to be manipulated in specific fixed poses. These positions and orientations are achieved by structuring the environment using jigs and fixtures; the rate at which objects are presented is controlled by various types of indexing systems. The migration towards flexible autonomous manufacturing systems necessitates the ability to reliably detect, recognize and continuously track objects within the workspace of a robot. The availability of such real-time sensory information, combined with recent advances in robot control, will enable a robot or other automated system to operate autonomously in a semi-structured or unstructured environment. The ultimate aim is to enable a robot system to operate with the same degree of flexibility as possessed by a human worker.