Intelligent Scaling Control for Internet-Based Teleoperation
Peter Xiaoping Liu*, Max Q-H Meng**, and Jason J. Gu***
*Department of Systems and Computer Engineering, Carleton University, Ottawa, ON, Canada K1S 5B6
**Department of Electronic Engineering, Chinese University of Hong Kong, Hong Kong
***Electrical and Computer Engineering, Dalhousie University, Halifax, Nova Scotia, Canada B3J 2X4
In this paper we present an intelligent scaling control algorithm, which is based on the hierarchical decomposition and dynamic assignment of tasks between the human operator and the on-board controller of the remote robot, for mobile robot teleoperation over the Internet. This control scheme relies on the real-time estimation of concurrent roundtrip delays in order to assign tasks between the user and the robot optimally. For this purpose, we employ a linear neural network for which most conventional learning algorithms are infeasible since their required computation is usually too intensive to be practical. To get over this dilemma, we introduce a novel learning algorithm based on the maximum entropy principle. Compared to traditional schemes, the computing cost of this algorithm is very low, which makes it possible for the proposed neural network to be implemented on-line in real time. The scaling control scheme with the neural-network-based delay prediction algorithm is successfully implemented and tested on the developed platform for Internet mobile robots.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 Internationa License.