TOWARDS AUTOMATIC VISUAL GUIDANCE OF AEROSPACE VEHICLES: FROM INSECTS TO ROBOTS

摘要

Equipped with a less-than-one-milligram brain, insects fly autonomously in complex environments without resorting to any Radars, Ladars, Sonars or GPS. The knowledge gained during the last decades on insects’ sensory-motor abilities and the neuronal substrates involved has provided us with a rich source of inspiration for designing tomorrow’s selfguided vehicles and micro-vehicles, which are to cope with unforeseen events on the ground, in the air, under water or in space. Insects have been in the business of sensory-motor integration for more than 100 millions years. They can teach us useful tricks for designing agile autonomous vehicles at various scales. Constructing a “biorobot” first requires exactly formulating the signal processing principles at work in the animal. It gives us, in return, a unique opportunity of checking the soundness and robustness of those principles by bringing them face to face with the real physical world. Here we describe some of the visually-guided terrestrial and aerial robots we have developed on the basis of our biological findings. All these robots react to the optic flow (i.e., the angular speed of the retinal image). Optic flow is sensed onboard the robots by miniature vision sensors called Elementary Motion Detectors (EMDs). The principle of these electro-optical velocity sensors was derived from optical/electrophysiological studies where we recorded the responses of single neurons to optical stimulation of single photoreceptor cells in a model visual system: the fly’s compound eye. Optic flow based sensors rely solely on contrast provided by reflected (or scattered) sunlight from any kind of celestial bodies in a given spectral range. These passive sensors and systems offer potential applications to manned or unmanned spacecraft, from robotic landers and rovers to asteroid explorers and satellite docking, with interesting prospects in weightreduction and low consumption.