Optimal Path Planning using an Improved A* Algorithm for Homeland Security Applications

A. Veeraswamy and B. Amavasai (UK)


Robotics, artificial intelligence, path planning, homeland security, search algorithms, sensor networks.


Path finding is an important sub task in mobile robotics and Homeland Security applications and has been subjected to extensive research. This paper analyses a variety of search algorithms used for path finding and planning. Using the Breve simulation environment, a general search algorithm and then the A* algorithm have been implemented. An improvement to the A* algorithm is introduced and presented. In this paper it has been proved through experimental results that the performance of the A* algorithm improves considerably after adding an additional heuristic. Dynamic path planning has also been implemented in this paper by allowing the vehicle to check for changes in the environment at every simulation time step and recalculate paths if there is a change in the environment. In the past ad-hoc sensor networks have been used in homeland security. In this paper ad-hoc sensor networks have been modeled using the patch class in the Breve simulation tool and path finding techniques have been used in this environment. Time and space complexity analysis of the various algorithms implemented in this paper have been presented.

Important Links:

Go Back