Roach Inspired Technology - February 8, 2005

To survive, animals must integrate large amounts of sensory information, often in milliseconds, to direct movement towards goals and away from threats. Similar directives are faced by autonomous vehicles through rapid assimilation of surroundings, instantaneous goal seeking and crash avoidance. Nature, through evolution and natural selection, has optimized this behavior. In particular, insects initiate actions reflexively in situations where no time for detailed information processing or planned decision-making is possible. This presentation will highlight innovative control strategies to maneuver autonomous vehicles using controls inspired by nature. Imagine the performance, speed and accuracy of autonomous vehicles possessing similar maneuvering capabilities of insects.

Members of the Orbital Research team have developed a biologically inspired neural network that mimics the escape response of the American Cockroach as well as swarming algorithms to coordinate a large number of manned, semi-autonomous and/or autonomous vehicles. These control strategies would empower vehicles with instantaneous collision avoidance as well as task allocation and optimization scenarios. These algorithms have been demonstrated on air vehicles demonstrating instantaneous reflexive obstacle avoidance response and advanced sensor integration.