A Self-Organizing Neural Network Architecture for Navigation Using Optic Flow

This article describes a self-organizing neural network architecture that transforms optic flow and eye position information into representations of heading, scene depth, and moving object locations. These representations are used to navigate reactively in simulations involving obstacle avoidance and pursuit of a moving target. The network's weights are trained during an action-perception cycle in which self-generated eye and body movements produce optic flow information, thus allowing the network to tune itself without requiring explicit knowledge of sensor geometry. The confounding effect of eye movement during translation is suppressed by learning the relationship between eye movement outflow commands and the optic flow signals that they induce. The remaining optic flow field is due to only observer translation and independent motion of objects in the scene. A self-organizing feature map categorizes normalized translational flow patterns, thereby creating a map of cells that code heading directions. Heading information is then recombined with translational flow patterns in two different ways to form maps of scene depth and moving object locations. Most of the learning processes take place concurrently and evolve through unsupervised learning. Mapping the learned heading representations onto heading labels or motor commands requires additional structure. Simulations of the network verify its performance using both noise-free and noisy optic flow information.

[1]  E Brenner,et al.  Detecting Changes in One's Own Velocity from the Optic Flow , 1994, Perception.

[2]  M. Banks,et al.  Perceiving heading with different retinal regions and types of optic flow , 1993, Perception & psychophysics.

[3]  H. C. Longuet-Higgins,et al.  A computer algorithm for reconstructing a scene from two projections , 1981, Nature.

[4]  A. V. D. Berg,et al.  Robustness of perception of heading from optic flow , 1992, Vision Research.

[5]  John Scott Bridle,et al.  Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition , 1989, NATO Neurocomputing.

[6]  W. H. Warren The state of flow , 1998 .

[7]  Olivier Faugeras,et al.  Self-calibration of a camera using multiple images , 1992, [1992] Proceedings. 11th IAPR International Conference on Pattern Recognition.

[8]  W. Simpson Optic flow and depth perception. , 1993, Spatial vision.

[9]  Nicholas G. Hatsopoulos,et al.  Visual navigation with a neural network , 1991, Neural Networks.

[10]  Stephen Grossberg,et al.  Vector associative maps: Unsupervised real-time error-based learning and control of movement trajectories , 1991, Neural Networks.

[11]  Constance S. Royden,et al.  Analysis of misperceived observer motion during simulated eye rotations , 1994, Vision Research.

[12]  H. Rodman,et al.  Coding of visual stimulus velocity in area MT of the macaque , 1987, Vision Research.

[13]  J. Lopez-Coronado,et al.  A real-time unsupervised neural network for the control of a mobile robot , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[14]  Robert J. Holt,et al.  Camera calibration problem: Some new results , 1991, CVGIP Image Underst..

[15]  A. V. van den Berg Perception of heading , 1993, Nature.

[16]  James A. Crowell,et al.  Estimating heading during real and simulated eye movements , 1996, Vision Research.

[17]  Leslie G. Ungerleider,et al.  Cortical connections of visual area MT in the macaque , 1986, The Journal of comparative neurology.

[18]  G. Orban,et al.  The Speed Tuning of Medial Superior Temporal (Mst) Cell Responses to Optic-Flow Components , 1995, Perception.

[19]  E. Brenner Judging object motion during smooth pursuit eye movements: The role of optic flow , 1991, Vision Research.

[20]  G. Wasilkowski,et al.  Computing optical flow , 1989, [1989] Proceedings. Workshop on Visual Motion.

[21]  K. Tanaka,et al.  Analysis of motion of the visual field by direction, expansion/contraction, and rotation cells clustered in the dorsal part of the medial superior temporal area of the macaque monkey. , 1989, Journal of neurophysiology.

[22]  Jacques Droulez,et al.  Stereo-motion cooperation and the use ofmotion disparity in the visual perception of 3-D structure , 1993, Perception & psychophysics.

[23]  R. Andersen,et al.  Mechanisms of Heading Perception in Primate Visual Cortex , 1996, Science.

[24]  C. Koch,et al.  Recurrent excitation in neocortical circuits , 1995, Science.

[25]  James A. Crowell,et al.  Estimating heading during eye movements , 1994, Vision Research.

[26]  Stefan Treue,et al.  Structure-from-motion: Perceptual evidence for surface interpolation , 1995, Vision Research.

[27]  Stephen Grossberg,et al.  Neural dynamics of adaptive sensory-motor control , 1986 .

[28]  R. Hetherington The Perception of the Visual World , 1952 .

[29]  W. Warren,et al.  The role of central and peripheral vision in perceiving the direction of self-motion , 1992, Perception & psychophysics.

[30]  Erkki Oja,et al.  Self-organizing maps for visually guided collision-free navigation , 1993, Proceedings of 1993 International Conference on Neural Networks (IJCNN-93-Nagoya, Japan).

[31]  Stephen Grossberg,et al.  Neural dynamics of motion grouping: from aperture ambiguity to object speed and direction , 1997 .

[32]  William H. Warren,et al.  Chapter 8 – Self-Motion: Visual Perception and Visual Control , 1995 .

[33]  Andrea J. van Doorn,et al.  Invariant Properties of the Motion Parallax Field due to the Movement of Rigid Bodies Relative to an Observer , 1975 .

[34]  John H. R. Maunsell,et al.  The connections of the middle temporal visual area (MT) and their relationship to a cortical hierarchy in the macaque monkey , 1983, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[35]  D Regan,et al.  How do we avoid confounding the direction we are looking and the direction we are moving? , 1982, Science.

[36]  Takeo Watanabe,et al.  High-Level Motion Processing , 1998 .

[37]  Ellen C. Hildreth,et al.  The perceptual buildup of three-dimensional structure from motion , 1989, Perception & psychophysics.

[38]  Juan López Coronado,et al.  An unsupervised neural network for low-level control of a wheeled mobile robot: noise resistance, stability, and hardware implementation , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[39]  J A Perrone,et al.  Model for the computation of self-motion in biological systems. , 1992, Journal of the Optical Society of America. A, Optics and image science.

[40]  Allan D. Jepson,et al.  Visual Perception of Three-Dimensional Motion , 1990, Neural Computation.

[41]  James A. Crowell,et al.  The perception of heading during eye movements , 1992, Nature.

[42]  J H Rieger,et al.  Processing differential image motion. , 1985, Journal of the Optical Society of America. A, Optics and image science.

[43]  G. Orban,et al.  Responses of macaque STS neurons to optic flow components: a comparison of areas MT and MST. , 1994, Journal of neurophysiology.

[44]  L. Dron Dynamic camera self-calibration from controlled motion sequences , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[45]  Ennio Mingolla,et al.  Global induced motion and visual stability in an optic flow illusion , 1997, Vision Research.

[46]  Stephen Grossberg,et al.  Contour Enhancement, Short Term Memory, and Constancies in Reverberating Neural Networks , 1973 .

[47]  S. Grossberg,et al.  Neural dynamics of planned arm movements: emergent invariants and speed-accuracy properties during trajectory formation. , 1988, Psychological review.

[48]  J. Rauschecker,et al.  Centrifugal motion bias in the cat's lateral suprasylvian visual cortex is independent of early flow field exposure. , 1990, The Journal of physiology.

[49]  Ellen C. Hildreth,et al.  Recovering heading for visually-guided navigation , 1992, Vision Research.

[50]  Helge J. Ritter,et al.  Kommunizierende Agenten: Gestische und natürlichsprachliche Interaktion , 1999, Kognitionswissenschaft.

[51]  Markus Lappe,et al.  A Neural Network for the Processing of Optic Flow from Ego-Motion in Man and Higher Mammals , 1993, Neural Computation.

[52]  D J Hannon,et al.  Eye movements and optical flow. , 1990, Journal of the Optical Society of America. A, Optics and image science.

[53]  H. C. Longuet-Higgins,et al.  The interpretation of a moving retinal image , 1980, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[54]  W. Simpson,et al.  Depth Discrimination from Optic Flow , 1988, Perception.

[55]  Helge J. Ritter,et al.  Ein hybrides Bildanalyse-System für einen künstlichen Kommunikator , 1999, Kognitionswissenschaft.

[56]  Christof Koch,et al.  Computing Optical Flow in the Primate Visual System , 1989, Neural Computation.

[57]  Allen M. Waxman,et al.  Surface Structure and Three-Dimensional Motion from Image Flow Kinematics , 1985 .

[58]  Daniel J. Hannon,et al.  Direction of self-motion is perceived from optical flow , 1988, Nature.

[59]  C. von der Malsburg Self-organization of orientation sensitive cells in the striate cortex. , 1973, Kybernetik.