Difference between revisions of "NSF CPS Project"
From Autonomous Systems Laboratory Wiki
| (One intermediate revision by one other user not shown) | |||
| Line 1: | Line 1: | ||
| − | [[Category: | + | [[Category:skynet]] |
=Overview= | =Overview= | ||
<!-- The purpose of the NSF Cyber-Physical Systems (CPS) research project is to explore high level methods and algorithms for high level autonomy. The areas of exploration include developing representations for constructing and maintaining models of complex environments, developing methods to anticipate changes in the environment and use them as part of planning, developing task planning methods that provide probabilistic guarantees for high-level behaviors --> | <!-- The purpose of the NSF Cyber-Physical Systems (CPS) research project is to explore high level methods and algorithms for high level autonomy. The areas of exploration include developing representations for constructing and maintaining models of complex environments, developing methods to anticipate changes in the environment and use them as part of planning, developing task planning methods that provide probabilistic guarantees for high-level behaviors --> | ||
Latest revision as of 12:24, 3 July 2013
Contents
Overview
The research goal of our NSF Cyber-Physical Systems (CPS) project is to develop algorithms for intelligent robotics which tightly integrate probabilistic perception and deterministic planning in a verifiable framework. Current activities include: 1) Developing representations for constructing and maintaining models of complex environments; 2) Developing methods to anticipate changes in the environment and use them as part of planning; 3) Developing task planning methods that provide probabilistic guarantees for high-level behaviors. The educational goal is to educate students in interdisciplinary CS and engineering projects and courses, particularly those related to intelligent robotics.
Research Projects
Multi-Path Planning
Synthesized controllers and obstacle prediction
Environment reconstruction
Publications Related to this project
- B. Johnson. F. Havlak, M. Campbell. H. Kress-Gazit, “Execution and Analysis of High-Level Tasks with Dynamic Obstacle Anticipation,” 2012 International Conference on Robotics and Automation.
- J. Hardy, M. Campbell, “Clustering Obstacle Predictions to Improve Contingency Planning for Autonomous Road Vehicles in Congested Environments,” IEEE/RSJ International Conference on Intelligent Robots and Systems, Sept 2011.
- M. Campbell, “Intelligent Autonomy in Robotic Systems,” The Bridge, Quarterly of the National Academy of Engineering, Winter 2010-11, pp 27-34.
- C. Rivadeneyra, M. Campbell, “Probabilistic Multi-Level Maps from LIDAR Data,” International Journal of Robotics Research, published on-line Jan 2011, Vol. 30 No. 12, Oct 2011, pp 1508-1526.
- M. Campbell, M. Egerstedt, J. How, R. Murray, “Autonomous Driving in Urban Environments: Approaches, Lessons and Challenges,” Philosophical Transactions of the Royal Society - A, Vol 368, 2010, pp. 4649-4672
- J. Schoenberg, A. Nathan, M. Campbell, “Segmentation of Dense Range Information in Complex Urban Scenes,” IEEE/RSJ International Conference on Intelligent Robots and Systems, Oct 2010.
- B. Johnson and H. Kress-Gazit, Probabilistic Analysis of Correctness of High-Level Robot Behavior with Sensor Error, In the Proceedings of Robotics: Science and Systems 2011, Los Angeles, CA, June 2011.
- I. Miller, M. Campbell, “Sensitivity Study of a Tightly-Coupled GPS / INS System for Autonomous Navigation,” IEEE Transactions on Aerospace and Electronic Systems, published on-line Dec 2011, Vol 48, No 2, April 2012, pp 1115-1135.
- I. Miller, M. Campbell, D. Huttenlocher, “Efficient Unbiased Tracking of Multiple Dynamic Obstacles Under Large Viewpoint Changes,” IEEE Transactions on Robotics, Vol 27, No 1, Feb 2011, pp 29-46.
- I. Miller, M. Campbell, D. Huttenlocher, et al, “Team Cornell's Skynet: Robust Perception and Planning in an Urban Environment,” Journal of Field Robotics, Vol 25, No 8, pp 493-527, 2008.
Funding
This material is based upon work supported by the National Science Foundation under Grant No. CNS-0931686
