Topic Num: A08-204 (Army)
Title: Multi-Robot Pursuit System
Research & Technical Areas: Ground/Sea Vehicles
Objective: Develop a software and sensor package to enable a team of robots to search for and detect human presence in an indoor environment.
Description: There are many research efforts within robotics in path planning, exploration, and mapping of indoor and outdoor environments. Operator control units are available that allow semi-autonomous map-based control of a team of robots. While the test environments are usually benign, they are slowly becoming longer and more complex. There has also been significant research in the game theory community involving pursuit/evasion scenarios. This topic seeks to merge these research areas and develop a software/hardware suit that would enable a multi-robot team, together with a human operator, to search for and detect a non-cooperative human subject. The main research task will involve determining the movements of the robot team through the environment to maximize the opportunity to find the subject, while minimizing the chances of missing the subject. If the operator is an active member of the search team, the software should minimize the chance that the operator may encounter the subject. As a simplification, the building layout could be given, although operating in an unknown environment with unknown obstacles is more realistic. The latter case should be studied at least in simulation. The software should maintain awareness of line-of-sight, as well as communication and sensor limits. It will be necessary to determine an appropriate sensor suite that can reliably detect human presence and is suitable for implementation on small robotic platforms. Additionally, the robot may not have the intelligence, sensing, or manipulative power to perform reconnaissance under full autonomy. For example, the robot may not be able to negotiate all obstacles, determine the course of action when confronted with difficult choices, or have sufficient team members to optimally search. Part of the research will involve determining what role the human operator will play in the search task. The system should flag the operator when assistance is required. Typical robots for this type of activity are expected to weigh less than 100 Kg and the team would have three to five robots.
PHASE I: Develop the system design and determine the required capabilities of the platforms and sensors. Perform initial feasibility experiments, either in simulation or with existing hardware. Documentation of design tradeoffs and feasibility analysis shall be required in the final report.
PHASE II: Implement the software and hardware into a sensor package, integrate the package with a generic mobile robot, and demonstrate the system’s performance in a suitable indoor environment. Deliverables shall include the prototype system and a final report, which shall contain documentation of all activities in this project and a user’s guide and technical specifications for the prototype system.
PHASE III: Robots that can intelligently and autonomously search for objects have potential commercialization within search and rescue, fire fighting, reconnaissance, and automated biological, chemical and radiation sensing with mobile platforms.
References: 1. http://carmen.sourceforge.net/home.html (Carnegie Mellon Robot Navigation Toolkit).
2. http:// www.informatik.uni-freiburg.de/~stachnis/pdf/pfaff07irosws.pdf (Navigation in Combined Outdoor and Indoor Environments using Multi-Level Surface Maps).
3. http://cis.jhu.edu/~rvidal/publications/tra01-final.pdf (Probabilistic Pursuit-Evasion Games: Theory, Implementation and Experimental Evaluation).
4. http://www-leibniz.imag.fr/perso/a0/fiorino/public_html/publications/aamas05.ps (Coordinated exploration of unknown labyrinthine environments applied to the Pursuit-Evasion problem).
5. https://drum.umd.edu/dspace/bitstream/1903/7085/1/TR+2007-13+Gehrels.pdf (Pursuit Techniques on Pioneer Robots).