Goal-directed navigation

(Last update: 05/Oct/2000)
Abstract:
While mobile robots and walking insects can use proprioceptive information (specialized receptors in the insect's leg, or wheel encoders in robots) to estimate distance traveled, flying agents have to rely mainly on visual cues. Experiments with bees provide evidence that flying insects might be using optical flow induced by egomotion to estimate distance traveled. Recently some details of this "odometer" have been unraveled. In this study, we propose a biologically inspired model of the bee's visual "odometer" based on Elementary Motion Detectors (EMDs), and present results from goal-directed navigation experiments with an autonomous flying robot platform that we developed specifically for this purpose. The robot is equipped with a panoramic vision system, which is used to provide input to the EMDs of the left and right visual fields. The outputs of the EMDs are in later stage spatially integrated by wide field motion detectors, and their accumulated response is directly used for the odometer. In a set of initial experiments, the robot moves through a corridor on a fixed route, and the outputs of EMDs, the odometer, are recorded. The results show that the proposed model can be used to provide an estimate of the distance traveled, but the performance depends on the route the robot follows, something which is biologically plausible since natural insects tend to adopt a fixed route during foraging. Given these results, we
 assumed that the optomotor response plays an important role in the context of goal-directed navigation, and we conducted experiments with an autonomous freely flying robot. The experiments demonstrate that this computationally cheap mechanism can be successfully employed in natural indoor environments.

Keywords: 3-D navigation, Visual odometer, Elementary Motion Detector, Flying robots, Biorobotics

Figure 1. Left: the Reichardt model of elementary motion detection. Photoreceptors, high-pass filters, low-pass filters, multipliers, and the subtraction module are wired in series. The output of the last step (subtraction) is an estimate of image speed (see text for details). Right: the visual odometer based on a wide field motion detector. The integrated response of an array of EMDs, each tuned to a different part of the visual field, is used to implement a wide field motion detector. Its outputs are then integrated over time to provide an estimate of the distance traveled (for explanations, see text).


 
 


Figure 2. Vision system and initial experiment setup. Left: The panoramic vision system we developed for this experiment consists of CCD camera, panoramic mirror, and housing components, which weighs 90 g in total so that it can be embedded in the flying robot. The visual field covers 360 degrees on the horizontal plane, and 260 degrees vertically. Middle and right: This vision system is equipped with the caster-frame on a pair of rails, and an external motor drives the vision system along a straight route at a constant speed. Initially walls with black and white stripes were installed along the rails (20cm width period of black and white, 60cm distance from vision system). 
 


Figure 10. Left: 3-D trajectories of the flying robot during the experiments. Each plot is extracted from the images recorded with the stereo video camera. Because of the same initial conditions, 5 trials show similar trajectories in both X-Y and Y-Z planes. Right: Visual odometer responses for the 5 trials. The similar trajectories of the robot induce the similar visual odometer measurements.
 

Reference
 

Send questions and comments to Fumiya Iida