A Synthetic Vision System Using Directionally Selective Motion Detectors to Recognize Collision

Reliably recognizing objects approaching on a collision course is extremely important. A synthetic vision system is proposed to tackle the problem of collision recognition in dynamic environments. The system combines the outputs of four whole-field motion-detecting neurons, each receiving inputs from a network of neurons employing asymmetric lateral inhibition to suppress their responses to one direction of motion. An evolutionary algorithm is then used to adjust the weights between the four motion-detecting neurons to tune the system to detect collisions in two test environments. To do this, a population of agents, each representing a proposed synthetic visual system, either were shown images generated by a mobile Khepera robot navigating in a simplified laboratory environment or were shown images videoed outdoors from a moving vehicle. The agents had to cope with the local environment correctly in order to survive. After 400 generations, the best agent recognized imminent collisions reliably in the familiar environment where it had evolved. However, when the environment was swapped, only the agent evolved to cope in the robotic environment still signaled collision reliably. This study suggests that whole-field direction-selective neurons, with selectivity based on asymmetric lateral inhibition, can be organized into a synthetic vision system, which can then be adapted to play an important role in collision detection in complex dynamic scenes.

[1]  Shigang Yue,et al.  Collision detection in complex dynamic scenes using an LGMD-based visual neural network with feature enhancement , 2006, IEEE Transactions on Neural Networks.

[2]  Shigang Yue,et al.  Visual motion pattern extraction and fusion for collision detection in complex dynamic scenes , 2006, Comput. Vis. Image Underst..

[3]  Georg von Wichert.  Can robots learn to see , 1999 .

[4]  F. Barth,et al.  Sensors and Sensing in Biology and Engineering , 2003, Springer Vienna.

[5]  Nicolas Franceschini,et al.  Visual guidance based on optic flow: a biorobotic approach , 2004, Journal of Physiology-Paris.

[6]  Jonathan A. Marshall,et al.  Self-organizing neural networks for perception of visual motion , 1990, Neural Networks.

[7]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[8]  B. Hassenstein,et al.  Systemtheoretische Analyse der Zeit-, Reihenfolgen- und Vorzeichenauswertung bei der Bewegungsperzeption des Rüsselkäfers Chlorophanus , 1956 .

[9]  Peter J. Fleming,et al.  The MATLAB genetic algorithm toolbox , 1995 .

[10]  José-Enrique Simó-Ten,et al.  Using infrared sensors for distance measurement in mobile robots , 2002, Robotics Auton. Syst..

[11]  Giacomo Indiveri,et al.  Neuromorphic vision sensors for mobile robots , 2001 .

[12]  Frank S. Werblin,et al.  Mechanisms and circuitry underlying directional selectivity in the retina , 2002, Nature.

[13]  Michael O'Shea,et al.  The Anatomy of a Locust Visual Interneurone; the Descending Contralateral Movement Detector , 1974 .

[14]  W. Taylor,et al.  Direction selectivity in the retina , 2002, Current Opinion in Neurobiology.

[15]  P. Simmons,et al.  Seeing what is coming: building collision-sensitive neurones , 1999, Trends in Neurosciences.

[16]  Christof Koch,et al.  A Silicon Implementation of the Fly's Optomotor Control System , 2000, Neural Computation.

[17]  Hobart R. Everett,et al.  Sensors for Mobile Robots: Theory and Application , 1995 .

[18]  F. Claire Rind,et al.  A DIRECTIONALLY SELECTIVE MOTION-DETECTING NEURONE IN THE BRAIN OF THE LOCUST: PHYSIOLOGICAL AND MORPHOLOGICAL CHARACTERIZATION , 1990 .

[19]  G. Schlotterer.  Response of the locust descending movement detector neuron to rapidly approaching and withdrawing visual stimuli , 1977 .

[20]  Margaret S Livingstone,et al.  Directional Inhibition A New Slant on an Old Question , 2005, Neuron.

[21]  Paul F. M. J. Verschure,et al.  Collision avoidance using a model of the locust LGMD neuron , 2000, Robotics Auton. Syst..

[22]  F Claire Rind,et al.  Motion detectors in the locust visual system: From biology to robot sensors , 2002, Microscopy research and technique.

[23]  Paul F. M. J. Verschure,et al.  Locust’s Looming Detectors for Robot Sensors , 2003 .

[24]  Barbara Webb,et al.  Reafferent or Redundant: Integration of Phonotaxis and Optomotor Behavior in Crickets and Robots , 2003, Adapt. Behav..

[25]  F. Rind,et al.  Neural network based on the input organization of an identified neuron signaling impending collision. , 1996, Journal of neurophysiology.

[26]  R D Santer,et al.  Retinally-generated saccadic suppression of a locust looming-detector neuron: investigations using a robot locust , 2004, Journal of The Royal Society Interface.

[27]  F C Rind,et al.  Orthopteran DCMD neuron: a reevaluation of responses to moving objects. II. Critical cues for detecting approaching objects. , 1992, Journal of neurophysiology.

[28]  Nicholas J. Priebe,et al.  Direction Selectivity of Excitation and Inhibition in Simple Cells of the Cat Primary Visual Cortex , 2005, Neuron.

[29]  Richard Stafford,et al.  A bio-inspired visual collision detection mechanism for cars: Optimisation of a model of a locust neuron to a novel environment , 2006, Neurocomputing.

[30]  H. Barlow,et al.  The mechanism of directionally selective units in rabbit's retina. , 1965, The Journal of physiology.

[31]  P. Simmons,et al.  Orthopteran DCMD neuron: a reevaluation of responses to moving objects. I. Selective responses to approaching objects. , 1992, Journal of neurophysiology.

[32]  Richard Stafford,et al.  A bio-inspired visual collision detection mechanism for cars: Combining insect inspired neurons to create a robust system , 2007, Biosyst..

[33]  Martin David Adams.  Sensor Modelling, Design and Data Processing for Autonomous Navigation , 1999, World Scientific Series in Robotics and Intelligent Systems.

[34]  George Adrian Horridge,et al.  What can engineers learn from insect vision? , 1992, Proceedings of IEEE Systems Man and Cybernetics Conference - SMC.

[35]  Heinrich H. Bülthoff,et al.  On robots and flies: Modeling the visual orientation behavior of flies , 1999, Robotics Auton. Syst..

[36]  Fumiya Iida,et al.  Biologically inspired visual odometer for navigation of a flying robot , 2003, Robotics Auton. Syst..

[37]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[38]  A. Borst,et al.  Neural networks in the cockpit of the fly , 2002, Journal of Comparative Physiology A.

[39]  Paul F. M. J. Verschure,et al.  Using a Mobile Robot to Study Locust Collision Avoidance Responses , 1999, Int. J. Neural Syst..

[40]  M. O'Shea,et al.  NEURONAL BASIS OF A SENSORY ANALYSER , THE ACRID ID MOVEMENT DETECTOR SYSTEM , 2005 .

[41]  Risto Miikkulainen,et al.  Modeling directional selectivity using self-organizing delay-adaptation maps , 2002, Neurocomputing.

[42]  Avinash C. Kak,et al.  Vision for Mobile Robot Navigation: A Survey , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[43]  H. Barlow,et al.  Selective Sensitivity to Direction of Movement in Ganglion Cells of the Rabbit Retina , 1963, Science.

[44]  Roberto Manduchi,et al.  Obstacle Detection and Terrain Classification for Autonomous Off-Road Navigation , 2005, Auton. Robots.

[45]  Shigang Yue,et al.  A Collision Detection System for a Mobile Robot Inspired by the Locust Visual System , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[46]  Richard H Masland,et al.  Functional inhibition in direction-selective retinal ganglion cells: spatiotemporal extent and intralaminar interactions. , 2002, Journal of neurophysiology.

[47]  F. Claire Rind,et al.  IDENTIFICATION OF DIRECTIONALLY SELECTIVE MOTION-DETECTING NEURONES IN THE LOCUST LOBULA AND THEIR SYNAPTIC CONNECTIONS WITH AN IDENTIFIED DESCENDING NEURONE , 1990 .