We have been developing a method for robot navigation based on panoramic vision. It is assumed that a robot has visited a number of positions in its workspace, and based on visual information only, it builds a topological map of the environment. This topological map is then used to drive the robot between positions in space.

• Contributions: Antonis Argyros, Costas Bekris.
• Current status: Method developed, simulations running.
• Sample results

Fig 1(a): Lefkos, the robotic platform of CVRL
with a panoramic camera mounted on it. It is used for experiments on robot homing.

Fig 1(b): Simulations of targeted navigation. Red color is walls,
red numbers are visited positions, blue numbers are features tracked,
black lines are permissible movements between locations, yellow trace is robot path from position 13 to position 2.
the topological map of the environment and the robot path planning
is performed automatically by the robot, using visual information only.

You may also download a video (2,81 Mbytes) showing the simulation running

Available documents:

• Technical report under development.