KEY PROBLEMS IN CREATING REAL-TIME EMBODIED AUTONOMOUS AGENTS.pptVIP

October 18-20, 2000 Learning to Behave -- Twente Workshop KEY PROBLEMS IN CREATING REAL-TIME EMBODIED AUTONOMOUS AGENTS Norman I. Badler Center for Human Modeling and Simulation University of Pennsylvania Philadelphia, PA 19104-6389 USA /~badler Outline Introduction and Applications Smart Agents Parameterized Action Representation PAR Agent Gestures Learning PARs Outline Introduction and Applications Smart Agents Parameterized Action Representation PAR Agent Gestures Learning PARs Applications for Embodied Human Agents: Engineering Design and Ergonomics Games Special Effects Military Simulations Job Education/Training Medical Simulations Embodied Human Agent “Dimensions” Appearance graphics Function human limits and abilities Time to animate movement Autonomy act in context Individuality generic or specific person Outline Introduction and Applications Smart Agents Parameterized Action Representation PAR Agent Gestures Learning PARs Why Smart Agents ? For Action Control GUI or 2D manipulation. Directly sensed 3D motion capture . Language commands text or speech . Use instructions -- as if the agent were a real person: A Smart Agent Smart Agent Requirements Actions to Execute: Action Representation - What it can do. Behavior Model: The agent’s decision-making, “thought,” and reaction processes - What it should do or wants to do. Inputs to Effect Behavior: Incoming knowledge about the outside world - What it needs to know. Agent Action Model 4 Levels of Action Representation 1: Basic Motion Generators 2: Parallel -Transition Networks 3: Parameterized Actions 4. Natural Language Instructions Level 1: Basic Human Movement Capabilities Gesture / Reach / Grasp / Carry. Walk / Turn / Climb. Posture Transitions Sit / Stand Visual Attention / Search. Motion playback captured or scripted . ‘Noise’ or secondary movements. Procedural Motions ? Economical Control Few parameters controlling many: Inverse kinematics