A novel potential field based domain mapping method

Authors:
Hari K. Voruganti;Bhaskar Dasgupta;Günter Hommel
Affiliations:
Department of Mechanical Engineering, Indian Institute of Technology Kanpur, India;Department of Mechanical Engineering, Indian Institute of Technology Kanpur, India;Technische Universität Berlin, Institut für Technische Informatik und Mikroelektronik, Berlin, Germany
Venue:
ICCOMP'06 Proceedings of the 10th WSEAS international conference on Computers
Year:
2006

Citing 4
Cited 1

Real-time obstacle avoidance for manipulators and mobile robots

International Journal of Robotics Research
Robot motion planning: a distributed representation approach

International Journal of Robotics Research
A comparative study of Dirichlet and Neumann conditions for path planning through harmonic functions

Future Generation Computer Systems - Special issue: Selected numerical algorithms
Soccer robot path planning based on the artificial potential field approach with simulated annealing

Robotica

Harmonic volumetric mapping for solid modeling applications

Proceedings of the 2007 ACM symposium on Solid and physical modeling

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new approach is proposed for mapping a geometrically complicated domain to a simple well-shaped domain using the potential field theory. This bijective domain mapping gives a parameterization which is useful in handling many application problems like path planning, shape matching, morphing, etc. Harmonic function is chosen for establishing the potential field as it will never have local minima within the domain. This paper presents the domain mapping method and an application of the same to the robot motion planning problem. Potential field along with the streamlines provides two parameters required for representation of any point in the domain. Once the domain is mapped, any post-processing like path planning is easy as the domain of operation is convex. for example, path planning boils down to finding a straight line joining two points in a convex domain and mapping it back to the original domain. Results show that domain mapping is an effective method for shape transformation. For on-line applications, this method is extremely useful since after mapping computational effort required is very less in the query phase.