RRT-Connect: An Efficient Approach to Single-Query Path Planning
ICRA, pp.995-1001, (2000)
A simple and efficient randomized algorithm is pre- sented for solving single-query path planning problems in high-dimensional configuration spaces. The method works by incrementally building two Rapidly-exploring Random Trees (RRTs) rooted at the start and the goal configurations. The trees each explore space around them and also advance...更多
下载 PDF 全文
- Motion planning problems arise in such diverse fields as robotics, assembly analysis, virtual prototyping, pharmaceutical drug design, manufacturing, and computer animation.
- Such problems involve searching the system configuration space of one or more complicated geometric bodies for a collision-free path that connects a given start and goal configuration, while satisfying constraints imposed by complicated obstacles.
- The key is to develop randomized methods that converge quickly in
- Motion planning problems arise in such diverse fields as robotics, assembly analysis, virtual prototyping, pharmaceutical drug design, manufacturing, and computer animation
- We present a simple path planning method called Rapidly-exploring Random Trees-Connect that combines Rapidly-exploring Random Trees (RRTs)  with a simple greedy heuristic that aggressively tries to connect two trees, one from the initial configuration and the other from the goal
- A randomized approach to single-query path planning is proposed that yields good experimental performance over a wide variety of examples
- The technique is based on Rapidly-exploring Random Trees (RRTs) and the Connect heuristic
- Some of the key practical advantages of the planning method include: 1) it does not require parameter tuning; 2) preprocessing is not required, yet interactive performance can be obtained for many difficult problems; 3) simple and consistent behavior was observed through repeated experiments; 4) a reasonable balance has been struck between greedy searching and uniform exploration; 5) the method is well-suited for incremental distance computation algorithms and fast nearest-neighbor algorithms
- Pathological cases exist for Rapidly-exploring Random Trees-Connect, and more experimental work is needed to determine conditions under which Rapidly-exploring Random Trees-Connect will yield very poor performance
- The authors present some preliminary experiments performed on a 270 MHz SGI O2 (R12000) workstation.
- Path smoothing was performed on the final paths to reduce jaggedness
- Some of these results are shown in Figure 7, in which the left column shows the RRTs, and the right column shows the corresponding solutions.
- The Connect heuristic only slightly increases running time in comparison to using the EXTEND function to construct two trees.
- It seems that the greedy behavior is worthwhile on average.
- The authors are currently comparing some of the variants discussed in Section 3
- The RRT-Connect has proven to be very successful in the experiments, the authors are aware of several intertwined factors that could improve performance even further.
- A randomized approach to single-query path planning is proposed that yields good experimental performance over a wide variety of examples.
- Some of the key practical advantages of the planning method include: 1) it does not require parameter tuning; 2) preprocessing is not required, yet interactive performance can be obtained for many difficult problems; 3) simple and consistent behavior was observed through repeated experiments; 4) a reasonable balance has been struck between greedy searching and uniform exploration; 5) the method is well-suited for incremental distance computation algorithms and fast nearest-neighbor algorithms.
- The practical performance observed so far is encouraging; an extensive study that involves many benchmarking examples would be useful, and is currently under investigation.
- Theoretical analysis of the convergence rate remains, which is one topic under current investigation
- Kuffner has been supported in part by a National Science Foundation Graduate Fellowship in Engineering, and MURI grant DAAH04-96-1-007
- LaValle has been supported in part by NSF CAREER Award IRI9875304 (LaValle)
- N. M. Amato and Y. Wu. A randomized roadmap method for path and manipulation planning. In IEEE Int. Conf. Robot. & Autom., pages 113–120, 1996.
- S. Arya, D. M. Mount, N. S. Netanyahu, R. Silverman, and A. Y. Wu. An optimal algorithm for approximate nearest neighbor searching. Journal of the ACM, 45:891–923, 1998.
- J. Barraquand, L.E. Kavraki, J.C. Latombe, T.Y. Li, R. Motwani, and P. Raghavan. A random sampling scheme for path planning. Int. J. Robot. Res., 16(6):759–774, 1997.
- J. Barraquand and J.-C. Latombe. Nonholonomic multibody mobile robots: Controllability and motion planning in the presence of obstacles. In IEEE Int. Conf. Robot. & Autom., pages 2328–2335, 1991.
- V. Boor, M. Overmars, and A.F. van der Stappen. The Gaussian sampling strategy for probabilistic roadmap planners. In Proc. of IEEE Int. Conf. Robotics and Automation, Detroit, MI, 1999.
- J. F. Canny. The Complexity of Robot Motion Planning. MIT Press, Cambridge, MA, 1988.
- D. Chalou and M. Gini. Parallel robot motion planning. In Proc. of IEEE Int. Conf. Robotics and Automation, pages 24–51, Atlanta, GA, 1993.
- S. Gottschalk, M. C. Lin, and D. Manocha. Obbtree: A hierarchical structure for rapid interference detection. In SIGGRAPH ’96 Proc., 1996.
- L. J. Guibas, D. Hsu, and L. Zhang. H-Walk: Hierarchical distance computation for moving convex bodies. In Proc. ACM Symposium on Computational Geometry, pages 265–273, 1999.
- T. Horsch, F. Schwarz, and H. Tolle. Motion planning for many degrees of freedom: Random reflections at c-space obstacles. In Proc. of the IEEE Int. Conf. on Robotics and Automation (ICRA’94), pages 3318– 3323, San Diego, CA, April 1994.
- D. Hsu, J.-C. Latombe, and R. Motwani. Path planning in expansive configuration spaces. In Int. J. of Computational Geometry and Applications, 1997.
- Y. K. Hwang and N. Ahuja. A potential field approach to path planning. IEEE Trans. Robot. & Autom., 8(1):23–32, February 1992.
- P. Indyk and R. Motwani. Approximate nearest neighbors: Towards removing the curse of dimensionality. In Proceedings of the 30th Annual ACM Symposium on Theory of Computing, 1998.
- J. J. Kuffner Jr. Autonomous Agents for Real-time Animation. PhD thesis, Stanford University, 1999.
- L. E. Kavraki, P. Svestka, J.-C. Latombe, and M. H. Overmars. Probabilistic roadmaps for path planning in high-dimensional configuration spaces. IEEE Trans. Robot. & Autom., 12(4):566–580, June 1996.
- L.E. Kavraki and J.-C. Latombe. Randomized preprocessing of configuration space for fast path planning. Technical report, Dept. of Computer Science, Stanford University, September 1993.
- Y. Koga, K. Kondo, J. Kuffner, and J.-C. Latombe. Planning motions with intentions. In Proc. SIGGRAPH ’94, pages 395–408, 1994.
- S. M. LaValle. Rapidly-exploring random trees: A new tool for path planning. TR 98-11, Computer Science Dept., Iowa State Univ. <http://janowiec.cs.iastate.edu/papers/rrt.ps>, Oct.1998.
- S. M. LaValle and J. J. Kuffner. Randomized kinodynamic planning. In Proc. IEEE Int’l Conf. on Robotics and Automation, 1999.
- M. C. Lin and J. F. Canny. Efficient algorithms for incremental distance computation. In IEEE Int. Conf. Robot. & Autom., 1991.
- E. Mazer, J. M. Ahuactzin, and P. Bessiere. The Ariadne’s clew algorithm. J. Artificial Intell. Res., 9:295– 316, November 1998.
- B. Mirtich. V-Clip: Fast and robust polyhedral collision detection. Technical Report TR97-05, Mitsubishi Electronics Research Laboratory, 1997.
- M. Overmars. A random approach to motion planning. Technical report, Dept. Computer Science, Utrect University, The Netherlands, October 1992.
- J. H. Reif. Complexity of the mover’s problem and generalizations. In Proc. of IEEE Symp. on Foundat. of Comp. Sci., pages 421–427, 1979.
- J. T. Schwartz and M. Sharir. On the piano movers’ problem: Coordinating the motion of several independent bodies. Int. J. Robot. Res., 2(3):97–140, 1983.
- P. Svestka. A probabilistic approach to motion planning for car-like robots. Technical report, Dept. Computer Science, Utrect Univ., April 1993.