OptiOdom: Circular Calibration Path for Omnidirectional Robots

0
17

Circular Calibration Path for Omnidirectional Robots

This video presents the circular calibration path used in OptiOdom’s experiments for omnidirectional robots. The circular calibration path is parameterized as follows:
– Initial radius – R0 (m): initial curvature radius that defines the first 180º curve
– Radius decreasing ratio – r% (%): ratio that is used to update the current curvature radius Rk after each 180º curve (Rk = r% * Rk-1; equivalent to the robot’s velocity module |v|k = r% * |v|k-1, at the k 180º curve)
– Number of 180º curves – #180º
– Number of total runs – N

Note that the robot has gone through the calibration path not only in the XR/v direction but also on the YR/vn one for observability purposes of the kinematic parameters of omnidirectional robots. Also for observability purposes, half of the runs for each v and vn directions should be performed in the clockwise direction and the other half in the opposite direction, as illustrated in the video.

One interesting advantage of this calibration path is the possibility to increase #180º (to increase the total distance made by the robot) while having the path’s required area restricted by the initial radius (if r% is lower than 1).

In the literature, the original circular path (similar to the one illustrated in the video and used in OptiOdom’s experiments) was firstly proposed by Kallasi et al. for tricycle robots and later used by Galasso et al. on the Ackerman steering geometry.

Bibliographic references:
– Kallasi et al., “A novel calibration method for industrial AGVs”, doi: 10.1016/j.robot.2017.04.019
– Galasso et al., “Efficient calibration of four wheel industrial AGVs”, doi: 10.1016/j.rcim.2018.11.005

—————————————-

OptiOdom: A Generic Approach for Odometry Calibration of Wheeled Mobile Robots

Authors:
– Ricardo B. Sousa [orcid: 0000-0003-4537-5095]
– Marcelo R. Petry [orcid: 0000-0002-7023-8562]
– Paulo G. Costa [orcid: 0000-0002-4846-271X]
– António Paulo Moreira [orcid: 0000-0001-8573-3147]

Acknowledgments:
– CRIIS – Centre for Robotics in Industry and Intelligent Systems from INESC TEC – Institute for Systems and Computer Engineering, Technology and Science
– Faculty of Engineering, University of Porto (FEUP)

source

LEAVE A REPLY

Please enter your comment!
Please enter your name here

12 − eleven =