Top Banner
© 2002 Imagination Engines, Inc. THE BRAIN OF THE TRULY AUTONOMOUS UAV Stephen L. Thaler, Ph.D. President & CEO, Imagination Engines, Inc. ___________________________________________________________ 40 th Annual NDIA Air Targets, UAVs Range Operations Symposium NEURAL NETWORKS
20

Artificially intelligent autopilots/airplanes/drones/UAVs

Nov 18, 2014

Download

Technology

Vapula

The Brain of the Truly Autonomous UAV.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • 1. NEURAL NETWORKS THE BRAIN OF THE TRULY AUTONOMOUS UAV Stephen L. Thaler, Ph.D. President & CEO, Imagination Engines, Inc.___________________________________________________________ 40th Annual NDIA Air Targets, UAVs Range Operations Symposium 2002 Imagination Engines, Inc.
  • 2. UNIVERSE IS A COMPLEX NETWORK 2002 Imagination Engines, Inc.
  • 3. NEURAL NETS MODEL UNIVERSE Connection weights are tantamount to expansion coefficients within a curve fit. input pattern = xcan be anybasis function because of some initiating event. because something else happenedMLPs capturecomplex causalchains. something happened output pattern = FN(FN-1(F1(x))) in contrast to the usual F(x) = a 0F0(x) + a2F1(x) + + aNFN(x) 2002 Imagination Engines, Inc.
  • 4. MLPS* CAPTURE ENTITIES & CONNECTS * MLP = Multi-Layer Perceptron, the workhorse of artificial neural networks. bitmaps of faces and other objects = x Eye, nose, mouth, classification ear, etc. eye detectors nose detectors detecting layer colonies form. If eyes, nose,constraint layer(s) mouth, and ears are detected then face is present. BOOL isaFace(x) output patterns BOOL isaFoot(x) 2002 Imagination Engines, Inc.
  • 5. CREATIVITY MACHINE PARADIGM hopping synapticCM perturbations inputs clamped for context Imagination Stream of Engine Ideas (IE) feedback AlertEvaluation Associative of Ideas Center (AAC) outputs US05659666, 08/19/199, Device for the Autonomous Generation of Useful Information 2002 Imagination Engines, Inc.
  • 6. AUTONOMOUS WARHEAD ADAPTATION 2002 Imagination Engines, Inc.
  • 7. AUTONOMOUS MOTION PLANNINGCreativity Machine 30 degree of freedom mastered inimplemented cockroach 30 seconds. A 1,000+ degree ofsimulation freedom system represented in flight controls do not pose a problem. GPS coordinates, chemical or biological gradients, etc. All that need be supplied is a will toUser drags roach while proceed in a certain direction,simulation invents the toward a particular goal, asrequired legwork! represented by the cursor drag. 2002 Imagination Engines, Inc.
  • 8. SELF-TRAINING ANN OBJECT synapticSTANNO adjustments input patterns Synaptic UntrainedCorrections Neural Network feedback ANN-BasedEvaluation of Neural Network Error Trainer outputs US05845271, 12/01/1998, Non-Algorithmically implemented artificial neural networks and components thereof 2002 Imagination Engines, Inc.
  • 9. STANNO-BASED ZDATA GENERATOR 2002 Imagination Engines, Inc.
  • 10. MEMBERSHIP / ANOMALY DETECTIONIntact feed forward passage through auto-associativenet, trained upon some interrelated group of patterns,signifies membership within that group of patterns.Features key to Relationships group G between membership sensed identified in features tested hidden layers. against those of group G in G = {P1, P2, PN} output layer(s). P = P P G P G P P P P G novelty vector, P P = (x1, x2, x3, , xN) US05852816, 12/22/1998, Neural network based database scanning system 2002 Imagination Engines, Inc.
  • 11. BATTLE DAMAGE ASSESSMENT 2002 Imagination Engines, Inc.
  • 12. BATTLE DAMAGE ASSESSMENT 2002 Imagination Engines, Inc.
  • 13. AUTONOMOUS TARGET RECOGNITION 2002 Imagination Engines, Inc.
  • 14. AUTONOMOUS TARGETING yquiescent xy (x, y) xy (0,0) x 2002 Imagination Engines, Inc.
  • 15. AUTONOMOUS TARGETINGperturbed xy (x, y) (x, y) (x, y) xy (0,0) x 2002 Imagination Engines, Inc.
  • 16. AUTONOMOUS TARGETING yperturbed Membership xy (x, y) filter (x, y) (x, y) xy (0,0) x 2002 Imagination Engines, Inc.
  • 17. AUTONOMOUS TARGETING yperturbed Membership xy (x, y) filter (x, y) (x, y) xy (0,0) x US05852816, 12/22/1998, Neural network based database scanning system 2002 Imagination Engines, Inc.
  • 18. AUTONOMOUS TARGETING 2002 Imagination Engines, Inc.
  • 19. AUTONOMOUS TARGETINGObjective: Autonomously find the toy F4attention window robot held camera 2002 Imagination Engines, Inc.
  • 20. AUTONOMOUS, BRILLIANT UAVSCreativity Machines + STANNOs + Group/Anomaly FiltersAircraft / UAV Design. Let them design themselves.Web-based Logistical Support Let them search for their own components.Field Kit for Tailored Assembly. Let them recommend their own field configurations.Self-diagnosis.. Let them tell us when / if theyre ready to go.Flight Control / Dogfight / Egress. Let them fly themselves.Massively Parallel Sensor Integration. Let them fuse inputs and resolve ambiguities.Mission / Sortie PlanningLet them plan based on loosely posed objectives.Strategy / Tactics Let them improvise as battlefield evolves.Battle Damage Recovery.. Let them repair themselves in flight.Low Observables Adaptation Let them reconfigure themselves to evade.Autonomous Targeting.. Let them lock on without slow human judgment.Battle Damage Assessment.. Let them evaluate their effects on target and react.Legal Repercussions.. Let them be their own, instantaneous cyber-lawyer.Political/Philosophical Perception. Let them have similar motivating feelings. 2002 Imagination Engines, Inc.