Please wait, Page is Loading...

A sensor-based controller able to treat total image loss and to guarantee non-collision during a vision-based navigation task

By D. Folio and V. Cadenat

  • David Folio is with IRISA-INRIA, Campus de Beaulieu, 35042 Rennes Cedex, France. David.Folio(at)
  • Viviane Cadenat is with CNRS ; LAAS ; 7, avenue du Colonel Roche, F-31077 Toulouse, France and Universite de Toulouse ; UPS, INSA, INP, ISAE ; LAAS-CNRS : F-31077 Toulouse, France. cadenat(at)


In this paper, we consider the problem of executing a vision-based task in an unknown environment. During such a task, two unexpected events may occur: the image data loss due to a camera occlusion and the robot collision with an obstacle. We first propose a method allowing to compute the visual data when they are wholly lost, before addressing the obstacle avoidance problem. Then, we design a sensor-based control strategy to perform safely vision-based tasks despite complete loss of the image. Simulation and experimental results validate our work.
visual servoing, collision avoidance, image features estimation.



To validate our work, we have first realized numerous simulations using Matlab software. Therefore, we have simulated a mission whose objective is to position the camera relatively to a landmark made of n=9 points. Moreover, the environment has been cluttered with two obstacles which may occlude the camera or represent a danger for the mobile base. Let us notice that the sampling period has been defined as to be close to our real robot one, that is Tech=50ms.

Simulation results

Fig.1: Simulation results.


We have also implemented the proposed control law on our mobile robot SuperScout II. We have considered a vision-based navigation task which consists in positioning the embedded camera in front of a given landmark made of n=4 points. The environment is cluttered with one obstacle, and envelopes ξ, ξ0 and ξ+ are respectively located at d=40cm, d0=56cm and d+=70cm from the obstacle.

Experimental results

Fig.2: Experimental results.

  • Add This
  • Add
  • Get
  • Google