Multi-aspect path planning for enhanced ground combat simulation

• Published in: SSCI : 2017 IEEE Symposium Series on Computational Intelligence : November 27, 2017-December 1, 2017 pp. 1 - 8
• Main Authors: Tolt, Gustav, Hedstrom, Johan, Bruvoll, Solveig, Asprusten, Martin
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2017
• Subjects: Algorithm design and analysis; Cost function; Navigation; optimization; Path planning; simulation; terrain analysis; Vegetation mapping; Visualization
• DOI: 10.1109/SSCI.2017.8280886

This paper describes the results of a collaborative Modelling and Simulation project between Norway and Sweden, aiming at improving the effectiveness of simulation-based battalion staff training. We show how advanced terrain analysis functionality can be integrated in a simulation framework to improve the basic behavior of entities in military simulations. A novel technique for path planning is proposed that combines a set of heterogeneous weights into a cost function serving as input to the A∗ algorithm to find the optimal path. The weights represent different aspects of path planning in military contexts: speed, accessibility, visual concealment, acoustic signature, cover, etc. Their relative importance can be adjusted by changing their priority weights, thereby obtaining routes that are optimal from different perspectives. Different sets of priorities are then grouped into path profiles, serving as a high level interface for integrating path planning functionality in host applications, such as more advanced behavior models. Implementation details and integration of the developed components in an existing simulation framework are described, along with the outcome of experiments and feedback from subject matter experts.