16.412J/6.835 Intelligent Embedded SystemsOutlineCourse Objective 1Basic Method LecturesCourse Objective 2Course Objective 3Slide 7Agents and IntelligencePowerPoint PresentationReflex agent with stateGoal-oriented agentUtility-based agentSlide 13Immobile Robots: Intelligent Offices and Ubiquitous ComputingEcological Life Support For Mars ExplorationSlide 16Slide 17Slide 18Slide 19Slide 20Cooperative ExplorationMIT Model Based Embedded and Robotics Group Autonomous Vehicles TestbedRobotic VehiclesIndoor test rangeScenarioCryobot & HydrobotSlide 27Slide 28Slide 29Four launches in 7 monthsSlide 31Traditional spacecraft commandingHouston, We have a problem ...Slide 3416.412J/6.835 Intelligent Embedded SystemsProf. Brian WilliamsRm 37-381Rm [email protected] 11-12:30, Rm 33-418OutlineOutline•Course Objectives and Assignments•Types of Reasoning•Kinds of Intelligent Embedded Systems•A Case Study: Space Explorers•Course Objectives and Assignments•Types of Reasoning•Kinds of Intelligent Embedded Systems•A Case Study: Space ExplorersCourse Objective 1Course Objective 1•To understand fundamental methods for creating the major components of intelligent embedded systems.Accomplished by:First ten lectures on basic methods~ 5 problem sets during the first ten lectures to exercise basic understanding of methods.•To understand fundamental methods for creating the major components of intelligent embedded systems.Accomplished by:First ten lectures on basic methods~ 5 problem sets during the first ten lectures to exercise basic understanding of methods.PlanExecuteMonitor &DiagnosisBasic Method LecturesBasic Method Lectures•Decision Theoretic Planning•Reinforcement Learning•Partial Order Planning•Conditional Planning and Plan Execution•Propositional Logic and Inference•Model-based Diagnosis•Temporal Planning and Execution•Bayesian Inference and LearningMore Advanced:•Graph-based and Model-based Planning•Combining Hidden Markov Models and Symbolic Reasoning•Decision Theoretic Planning•Reinforcement Learning•Partial Order Planning•Conditional Planning and Plan Execution•Propositional Logic and Inference•Model-based Diagnosis•Temporal Planning and Execution•Bayesian Inference and LearningMore Advanced:•Graph-based and Model-based Planning•Combining Hidden Markov Models and Symbolic ReasoningCourse Objective 2Course Objective 2•To dive into the recent literature, and collectively synthesize, clearly explain and evaluate the state of the art in intelligent embedded systems.Accomplished by:Weekly thought questions (~ 2 page answers)Group lecture on advance topic45 minute lectureShort tutorial article on method 1-3 methodsDemo of example reasoning algorithmGroups of size ~3.•To dive into the recent literature, and collectively synthesize, clearly explain and evaluate the state of the art in intelligent embedded systems.Accomplished by:Weekly thought questions (~ 2 page answers)Group lecture on advance topic45 minute lectureShort tutorial article on method 1-3 methodsDemo of example reasoning algorithmGroups of size ~3.Course Objective 3Course Objective 3•To apply one or more reasoning elements to create a simple agent that is driven by Goals or RewardsAccomplished by:Final project during last third of courseImplement and demonstrate one or more reasoning methods on a simple embedded system.Short final presentation on project.Final project report.•To apply one or more reasoning elements to create a simple agent that is driven by Goals or RewardsAccomplished by:Final project during last third of courseImplement and demonstrate one or more reasoning methods on a simple embedded system.Short final presentation on project.Final project report.PlanExecuteMonitor &DiagnosisOutlineOutline•Course Objectives and Assignments•Types of Reasoning(Slides compliments of Prof Malik, Berkeley)•Kinds of Intelligent Embedded Systems•A Case Study: Space Explorers•Course Objectives and Assignments•Types of Reasoning(Slides compliments of Prof Malik, Berkeley)•Kinds of Intelligent Embedded Systems•A Case Study: Space ExplorersAgents and IntelligenceAgents and IntelligenceProf Malik, BerkeleyReflex agentsReflex agentsCompliments of Prof Malik, BerkeleyReflex agent with stateReflex agent with stateCompliments of Prof Malik, BerkeleyGoal-oriented agentGoal-oriented agentCompliments of Prof Malik, BerkeleyUtility-based agentUtility-based agentCompliments of Prof Malik, BerkeleyOutlineOutline•Course Objectives and Assignments•Types of Reasoning•Kinds of Intelligent Embedded Systems•A Case Study: Space Explorers•Course Objectives and Assignments•Types of Reasoning•Kinds of Intelligent Embedded Systems•A Case Study: Space ExplorersImmobile Robots: Intelligent Offices and Ubiquitous ComputingEcological Life SupportFor Mars Explorationcourtesy NASAThe MIR Failurecourtesy NASA AmesMIT Spherescourtesy Prof. Dave Miller, MIT Space Systems Laboratorycourtesy JPLDistributed Spacecraft Interferometers to Distributed Spacecraft Interferometers to search for Earth-like Planets Around Other Starssearch for Earth-like Planets Around Other Starscourtesy JPL``Our vision in NASA is to open the Space Frontier . . . We must establish a virtual presence, in space, on planets, in aircraft and spacecraft.’’ - Daniel S. Goldin, NASA Administrator, May 29, 1996A Goldin Era of Robotic Space ExplorationCooperative ExplorationDistributed Planning Group, JPLModel-based Embedded and Robotic Systems Group, MITMIT Model Based Embedded and Robotics Group Autonomous Vehicles TestbedMIT Model Based Embedded and Robotics Group Autonomous Vehicles TestbedRobotic VehiclesRobotic Vehicles•ATRV Rovers•Monster Trucks•Blimps•Spheres •Simulated Air/Space Vehicles•ATRV Rovers•Monster Trucks•Blimps•Spheres •Simulated Air/Space VehiclesIndoor test rangeIndoor test rangeAim & Scope:•indoor experiments for target site exploration•cooperative explorationScenarioScenarioCooperative Target Site Exploration:Heterogeneous rover team and blimps explore science sites determined by remote sensingexploration featurepath planned/takenway pointexploration regionidentified featuregoal positionTasks:•small scout rovers (ATRV Jr) explore terrain as described in earlier scenarios•blimps provide additional fine grain air surveillance•scout rovers identify features for further
View Full Document