A constraint-based solver for the military unit path finding problem

Authors:
Louise Leenen;Johannes Vorster;Hermanus le Roux
Affiliations:
Council for Industrial and Scientific Research, South Africa, Brummeria, Pretoria;Council for Industrial and Scientific Research, South Africa, Brummeria, Pretoria;Council for Industrial and Scientific Research, South Africa, Brummeria, Pretoria
Venue:
SpringSim '10 Proceedings of the 2010 Spring Simulation Multiconference
Year:
2010

Citing 6
Cited 1

Robot motion planning: a distributed representation approach

International Journal of Robotics Research
Artificial intelligence: a modern approach

Artificial intelligence: a modern approach
Constraint Processing

Constraint Processing
Incremental heuristic search in AI

AI Magazine
Balancing safety and speed in the military path finding problem: analysis of different ACO algorithms

Proceedings of the 9th annual conference companion on Genetic and evolutionary computation
Hierarchical A *: searching abstraction hierarchies efficiently

AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1

hCHAC: A family of MOACO algorithms for the resolution of the bi-criteria military unit pathfinding problem

Computers and Operations Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

We describe the first phase development of a path finding simulation in a military environment. This concept demonstrator can be used for mission planning by constructing what-if scenarios to investigate trade-offs such as location of deployment and mode of transport. The Military Unit Path Finding Problem (MUPFP) is the problem of finding a path from a starting point to a destination where a military unit has to move, or be moved, safely whilst avoiding threats and obstacles and minimising costs in a digital representation of the real terrain [1]. Although significant research has been done on path finding, the success of a particular technique relies on the environment and existing constraints. The MUPFP is ideal for a constraint-based approach because it requires flexibility in modelling. We formulate the MUPFP as a constraint satisfaction problem and a constraint-based extension of the A* search algorithm. The concept demonstrator uses a provided map, for example taken from Google Earth, on which various obstacles and threats can be manually marked. Our constraint-based approach to path finding allows for flexibility and ease of modelling. It has the advantage of modelling new environments or additional constraints with ease, and it produces near-optimal solutions if solving is halted prematurely.