Compact image processing based kin recognition, distance measurement and identification method in a robot swarm

Abstract

In this paper we develop a reliable visual recognition, distance measurement and identification algorithm to detect and identify kin robots in a mobile-robot swarm. Every robot in the swarm is equipped with a zebra pattern and our idea is based on the Fast Fourier Transform (FFT), which has a relatively low complexity. We sampled the camera images of the robot by a vertical sampling method, and the sampled one dimensional vectors were the inputs of the FFT. The basis of the pattern localization is that there is a distinguishable peak in the FFT spectrum if the sample went through a horizontal zebra pattern. If this peak is larger than an empirically determined threshold value the algorithm finds the location of a kin robot. Moreover, based on the FFT spectrum this method it is capable to estimate the distance of the kin robot. Besides, we modified the zebra pattern to a barcode like pattern, and so the identification of the swarm members is also possible. To test our idea several experiments were accomplished with Surveyor SRV-1 robots, which have an on-board camera system. The captured pictures of the observer robot were sent to and processed in a PC using MATLAB program.