Learning implicit models during target pursuit
Gaskett, Chris, Brown, Peter, Cheng, Gordon, and Zelinsky, Alexander (2003) Learning implicit models during target pursuit. In: IEEE international conference on robotics and automation (3), pp. 4122-4129. From: Proceedings of the IEEE international conference on robotics and automation (ICRA2003), 14-19 September 2003, Taiwan.
Smooth control using an active vision head’s verge-axis joint is performed through continuous state and action reinforcement learning. The system learns to perform visual servoing based on rewards given relative to tracking performance. The learned controller compensates for the velocity of the target and performs lag-free pursuit of a swinging target. By comparing controllers exposed to different environments we show that the controller is predicting the motion of the target by forming an implicit model of the target’s motion. Experimental results are presented that demonstrate the advantages and disadvantages of implicit modelling.
|Item Type:||Conference Item (Refereed Research Paper - E1)|
|Keywords:||reinforcement learning, active head, visual servoing|
|Date Deposited:||04 Oct 2006|
|FoR Codes:||08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080101 Adaptive Agents and Intelligent Robotics @ 0%|
Last 12 Months: 6