IIC 06 21 Real time mobile robot control with a multiresolution map representation Katsuya Iwata, Shinkichi Inagaki, Yusuke Nara, Tatsuya Suzuki (Nagoya University) Abstract In this paper a real-time path planning algorithm for a single mobile robot which has omni-directional vision sensors is proposed. The robot implements RTA* - real-time heuristic search algorithm, by which the robot can reach its goal position by alternating searching and moving. For this implementation, the robot needs to transform its sensory information to graph searching space,and to solve this, a method of the transformation for an omni-directional distance sensor is proposed. A multi-resolution searching architecture is also proposed which changes the sensor resolution according to the robot s facing situations, and by changing the moving speed of the robot simultaneously, the robot is controlled with a control request. In addition, probabilistic decision making algorithm is examined, by which the reaching time is decreased in some situations. Finally simulation results show the validities of the proposed methods. RTA*SLAM, (Mobile Robots, Navigation, Real Time Path Planning, RTA*, Multiresolution, SLAM, Omni-Directional Sensor ) 1. Simultaneous Localization and Map Building(SLAM ) Dissanayake (1) SLAM (6) Thrun (2) (3) Baum-Welch Cuperlier (4) DOG Shoudong (5) Real-Time A*(RTA*) RTA* 1/6
2 3 Real- Time A* 4 5,6 2. Fig.1 1 Fig. 1. Environment Example 3. RTA* 1) 2) A* Korf (7) Real-Time A*(RTA*) RTA* RTA* 1 2 RTA* Fig. 2. Cost Function of RTA* Search n d n + d f(n, n + d) n n + d (Fig.2) f(n, n + d) =g(n, n + d)+h(n + d, n G ). (1) g(n, n + d) n n + d h(n + d, n G) n + d n (x, y) n G (x G,y G ) h(n + d, n G )= (x G x) 2 +(y G y) 2. (2) 2 f(n, n + d) = min f(n, n + d) (3) n+d N n + d n n + d n +1 N n + d n h(n, n G) f(n, n+d) h(n, n G ) f(n, n ) h(n, n G) h(n, n G ) min f(n, n + d). (4) n+d N {n } RTA* 4. RTA* 2/6
(8) 1 1 2 4 1 Measurable Area: MA R MA MA (Fig.3) Rule1 : MA (Node Production Sector) Rule2 : MA (Node Production Point) Rule3 : MA R (Node Production Segment) Rule4 : MA (Production Prohibited Area) 4 2 (9) (10) (11) (12) MA R (6) (5) (8) Fig.5 STEP1 : (5) R STEP2 : (6) R θ (7) D(R, θ) =R R STEP3 : (8) STEP2 3 Fig. 3. Node Production Rules 4 Fig. 4. Parameters 3/6
R = R max. (5) 2π D(R, θ)dθ 0 R =, (6) 2π D(R, θ) <R. (7) 2π 2π Rdθ D(R, θ)dθ T hreshold. (8) 0 0 R MA R max MA θ D(R, θ) MA R θ MA (Fig.4) 5. RTA* 5 1 Fig.6 MA 5 2 MA (5)(6)(7)(8) Fig.7(c) Fig.7(a),(b) MA MA 6 MA Fig. 6. Path Planning Simulation with Fixed Measurable Area Radius Fig. 5. 5 MA Algorithm for Determination of MA Radius 7 MA Fig. 7. Variable Measurable Area Radius 4/6
6. 2 DC 2 2m/s (Fig.8) 360 Fig.9 Fig.10 Fig.10 (a) MA 800mm (b) MA 4000mm (c) MA MA 1120mm, 1280mm MA 720mm 9 Fig. 9. Environment for the Experiment 8 Fig. 8. Mobile Robots 10 Fig. 10. Built Maps with fixed and variable Mesurable Area Radius 7. RTA* 1 M.W.Gamini Dissanayake, Paul Newman, Steven Clark, Hugh F. Durrant-Whyte, and M.Csorba A Solution to the Simultaneous Localization and Map Building(SLAM) Problem IEEE Trans. on Robotics and Automation, Vol. 17, No. 3, pp. 229-241, 2001 2 Sebastian Thrun, Wolfram Burgard, and Dieter Fox A Probabilistic Approach to Concurrent Mapping and 5/6
Localization for Mobile Robots Machine Learning and Autonomous Robots, 1998 3 Sebastian Thrun Leaning metric-topological maps for indoor mobile robot navigation Artificial Intelligence, Vol. 99, No. 1, pp. 21-71, 1998 4 Nicholas Cuperlier, Mathias Quoy, Philippe Gaussier, and Christophe Giovanangelli Navigation and planning in an unknown environment using vision and a cognitive map IJCAI Workshop Reasoning with Uncertainty in Robotics, pp. 48-53, 2005 5 Shoudong Huang, Zhan Wang, and G.Dissanayake Time optimal robot motion control in simultaneous localization and map building(slam) problem 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2004.(IROS 2004). Proceedings., Vol.3, pp. 3110-3115, 2004 6, Tauseef Gulrez, Damith C. Herath, and Gamini Dissanayake ( ) (C ), Vol. 71, No. 703, pp. 1522-1526, 2005 7 R. Korf Real-Time Heuristic Search Artificial Intelligence, Vol. 42, No. 2 3, pp. 189 211, 1990 8 2001. 9 Gerhard K. Kraetzschmar, Guillem Pages Gassull, and Klaus Uhl Probabilistic Quadtrees for Variable- Resolution Mapping of Large Environments Proceedings of the 5th IFAC/EURON Symposium on Intelligent Autonomous Vehicles, 2004 10 S.Kambhampati, and L.Davis Multiresolution path planning for mobile robots IEEE Journal of Robotics and Automation, Vol. 2, pp 135-145, 1988 11,, and, Vol. 7, No. 5, pp. 403-413, 1989 12,, and, Vol. 7, No. 5, pp. 414-424, 1989 6/6