Autonomous Systems Development at AFRL June 25, 2002 Paul Paul Zetocha Zetocha Paul Zetocha Group Lead, Intelligent Satellite Systems Group Lead, Intelligent Satellite Systems Group Lead, Intelligent Satellite Systems AFRL/VS AFRL/VS AFRL/VS (505) 853- -4114 4114 (505) 853-4114 (505) 853 Paul.Zetocha Zetocha@ @kirtland kirtland. .af af.mil .mil Paul. Paul.Zetocha@kirtland.af.mil 1
Satellite Autonomy and Satellite Autonomy and Fault Detection & Recovery Fault Detection & Recovery Distributed systems require autonomous Distributed systems require autonomous decision-making among multiple satellites decision-making among multiple satellites Cluster Management Cluster Management Other Message Center Other Message Center Satellites Satellites • Distributed processing • Distributed processing • Agent-based communication • Agent-based communication Agent Message Center Agent • Formation flying process control • Formation flying process control Agent • Data fusion • Data fusion Spacecraft Agent • Virtual satellite command • Virtual satellite command Spacecraft and control and control Message Center Research in Smart Systems Research in Smart Systems Agent Ground Cluster geometry Cluster geometry Collision Collision formation and maintenance formation and maintenance Avoidance Avoidance – Flocking – Flocking – Goal directed – Goal directed behavior behavior behavior Cooperative behavior Cooperative Processing Processing Fault Detection Fault Detection Isolation & Resolution Isolation & Resolution Distributed Distributed – Genetic algorithms, – Genetic algorithms, Resource – State-based, rule-based, Resource – State-based, rule-based, neural networks, neural networks, Allocation Allocation case-based, and case-based, and fuzzy logic fuzzy logic – Market negotiation – Market negotiation model-based reasoning model-based reasoning 2
Cluster Manager Objectives On-Board Formation Flying Science Processing Process Control Fault Management • Change detection • Relative positioning • Feature recognition • On-board knowledge to cm level • Trigger to perform base • Configurations from reconfiguration • Limit Checking 100m to 5 Km • Autonomous data recollect • Real-time reaction • Collision avoidance • Optimization of science return • CM Rollover • SV mode maintenance Command & Telemetry • Virtual Satellite Control On-Board Planning • Cluster level knowledge • Intelligent SV replanning due maintained with SOH data to mission events passed through ISL • Real-time • Commanding to individual response satellites or to CM and then • Intelligent routed sensor • Consolidated telemetry queuing 3
ASE Mission Scenario Target Image with Onboard Science Cluster Management: Autonomous Onboard Processing and Constellation New Science Images Spacecraft Replanning Event Detection Reconfiguration Constellation 4
Information Fusion Information Fusion The Key is Seamless FLOW to the DECISION MAKER Decision-Specific Information • Timely • Intel Sources • Consistent Situation • Air Surveillance Weather • Structured • Surface Intelligence • Tailored Surveillance Coalition Forces • Space • High Quality Logistics Surveillance Imagery Overlays • Integrated Decision Maker Terrain/Cultural Features Fusion Technology DATA INFORMATION KNOWLEDGE UNDERSTANDING 5
How Supercomputing can Enable Autonomy Help transition to an era where we can accept and trust satellite autonomy • High fidelity simulations of the space environment to assist in the development, test, and evaluation of autonomous software • High fidelity simulations of anomaly scenarios 6
How Supercomputing can Enable Autonomy Enable Future Missions • Model-based reasoning and planning systems that depict satellite components to extremely high levels of detail and that can run anomaly resolution scenarios extremely fast • Virtual reality / 3-D representations of satellites that would allow an analyst to “step inside” and interact with individual components • Conduct exhaustive searches of the possible results of operations such as switching to a redundant string 7
Recommend
More recommend
