Kinematic modelling of mobile robots by vision based algorithm

This paper deals with a vision-based algorithm for kinematic modelling of mobile robots. Firstly an approach for tracking of mobile robots in a platooning syste m which calculates the distance between the robots using onl y a single digital camera with support of infra-red range finde rs is presented. The key features defining the distance b etween two robots are obtained by image processing. The calcul ated distances by the two sensor channels are synchronize d in time and fused with a weighted average algorithm. The cont rol of transport means driven in a platoon relies on linea r and angular velocity of the preceding transport means. Secondly an approach for angular speed defining by means of computer vis ion is presented. A linear segment is extracted from an im age in order to calculate the angle shift and the velocity of the robot using a simple geometrical relation. A comparison between th e proposed geometrical approach and the classical Look Up Tabl es (LUT) solution shows: i) that the relation between key feat ures and distance is nonlinear and the data from different s ources are reliably fused in a short computation time, ii) alt hough the LUT is more accurate, the proposed geometrical approach can be useful in transport means due to its simplicity and short computational time.

[1]  Scott Alan Crawford,et al.  Performance Evaluation of Sensor Combinations for Mobile Platoon Control , 2005 .

[2]  Tien Fang Fwa,et al.  Applications of Advanced Technologies in Transportation Engineering , 2004 .

[3]  Giuseppe Oriolo,et al.  Mobile robot localization via fusion of ultrasonic and inertial sensor , 2000 .

[4]  Markus Vincze,et al.  Dynamic measurement of position and orientation of robots , 1992 .

[5]  M.E. Khatir,et al.  Decentralized control of a large platoon of vehicles operating on a plane with steering dynamics , 2005, Proceedings of the 2005, American Control Conference, 2005..

[6]  Pravin Varaiya,et al.  Smart cars on smart roads: problems of control , 1991, IEEE Trans. Autom. Control..

[7]  José Santos-Victor,et al.  Vision-based navigation and environmental representations with an omnidirectional camera , 2000, IEEE Trans. Robotics Autom..

[8]  Michel Parent,et al.  Vision Technique for Platoon Driving , 1996 .

[9]  Man Hyung Lee,et al.  Localization of a mobile robot using images of a moving target , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[10]  G. Oriolo,et al.  MOBILE ROBOT LOCALIZATION VIA FUSION OF ULTRASONIC AND INERTIAL SENSOR DATA , 2000 .

[11]  Said Mammar,et al.  Sub-Optimal Longitudinal Control of Road Vehicles With Capacity and Safety Considerations , 2004 .

[12]  Ching-Chih Tsai,et al.  Laser pose tracking for a mobile robot using fuzzy adaptive extended information filtering , 2003, Proceedings of the 2003 American Control Conference, 2003..

[13]  Chia-Ju Wu,et al.  A vision-based method for localization of mobile robots , 2007 .

[14]  Tieniu Tan,et al.  Mobile robot self-localization based on global visual appearance features , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[15]  Chia-Ju Wu,et al.  Localization of an Autonomous Mobile Robot Based on Ultrasonic Sensory Information , 2001, J. Intell. Robotic Syst..

[16]  W H Schulz,et al.  COST-BENEFIT ANALYSES OF CHAUFFEUR SYSTEMS - METHODOLOGY AND EMPIRICAL RESULTS , 2001 .

[17]  Peter Mitrouchev,et al.  Visual-based system for tracking of mobile robots with support of infra-red sensor , 2006 .

[18]  Brahim Chaib-draa,et al.  Collaborative Driving System Using Teamwork for Platoon Formations , 2005, Applications of Agent Technology in Traffic and Transportation.

[19]  Henrik I. Christensen,et al.  Pose tracking using laser scanning and minimalistic environmental models , 2001, IEEE Trans. Robotics Autom..