A neural-network approach to high-precision docking of autonomous vehicles/platforms

Authors:
Joseph Wong;Goldie Nejat;Robert Fenton;Beno Benhabib
Affiliations:
Computer Integrated Manufacturing Laboratory, Department of Mechanical and Industrial Engineering University of Toronto, 5 King's College Road, Toronto, M5S 3G8, Ontario, Canada;Department of Mechanical Engineering, State University of New York at Stony Brook, Stony Brook, 11794–2300, New York, USA;Computer Integrated Manufacturing Laboratory, Department of Mechanical and Industrial Engineering University of Toronto, 5 King's College Road, Toronto, M5S 3G8, Ontario, Canada;Computer Integrated Manufacturing Laboratory, Department of Mechanical and Industrial Engineering University of Toronto, 5 King's College Road, Toronto, M5S 3G8, Ontario, Canada
Venue:
Robotica
Year:
2007

Citing 4
Cited 0

Original Contribution: A scaled conjugate gradient algorithm for fast supervised learning

Neural Networks
On-line learning in neural networks

On-line learning in neural networks
Correction of systematic odometry errors in mobile robots

IROS '95 Proceedings of the International Conference on Intelligent Robots and Systems-Volume 3 - Volume 3
A guidance-based motion-planning methodology for the docking of autonomous vehicles: Research Articles

Journal of Robotic Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a Neural-Network-(NN) based guidance methodology is proposed for the high-precision docking of autonomous vehicles/platforms. The novelty of the overall online motion-planning methodology is its applicability to cases that do not allow for the direct proximity measurement of the vehicle's pose (position and orientation). In such instances, a guidance technique that utilizes Line-of-Sight-(LOS) based task-space sensory feedback is needed to minimize the detrimental impact of accumulated systematic motion errors. Herein, the proposed NN-based guidance methodology is implemented during the final stage of the vehicle's motion (i.e., docking). Systematic motion errors, which are accumulated after a long-range motion are reduced iteratively by executing corrective motion commands generated by the NN until the vehicle achieves its desired pose within random noise limits. The proposed guidance methodology was successfully tested via simulations for a 6-dof (degree-of-freedom) vehicle and via experiments for a 3-dof high-precision planar platform.