Today s Assignment. Outline. Principles of Autonomy and Decision Making. Objectives and Logistics Appendix: Agents and Their Building Blocks

Transcription

1 Principles of Autonomy and Decision Making Brian Williams / Session 26b 1 Today s Assignment Study for the final! Outline Objectives and Logistics Appendix: Agents and Their Building Blocks 1

2 Course Objective 1: Agent Architectures 1. To appreciate the major types of agents, their major functions and the applications they support. 2. To understand the common architectures used to develop agents. Understanding exercised through case studies. Agent Architecture (Objective 1) Communicate and Interpret Locate in World Monitor & Diagnosis Plan Execute Map Plan Routes Maneuver and Track Types of Agents (Objective 1) Mission-oriented Agents Self-repairing Agents Mobile Agents Agile Agents Communicating Agents 2

3 Course Objective 2: Principles of Agents /13: To learn the modeling and algorithmic building blocks for creating reasoning and learning agents: 1. To formulate reasoning problems in an appropriate formal representation. 2. To describe, analyze and demonstrate the application of reasoning algorithms to solve these problem formulations. Understanding demonstrated on paper and through implementation. Introduction to modeling, algorithms and analysis the next two Wednesday. Introduction to implementation the next two Mondays. Agent Building Blocks Activity Planning Execution/Monitoring Diagnosis Repair Scheduling Resource Allocation Path Planning Localization Map Building Trajectory Design Policy Construction Course Objective 3: Implementing Agents : To appreciate the challenges of building a state of the art autonomous explorer: Fall 03: Mars Exploration Rover shadow mode demonstration. Fall 04: Gnu Robot competition. Fall 05: Model-based autonomy toolbox The virtual solar system Stay tuned for more. 3

4 Outline Objectives and Logistics Appendix: Agents and Their Building Blocks Types of Agents (objective 1) 1. Mission-oriented Agents 2. Self-repairing Agents 3. Mobile Agents 4. Agile Agents 5. Communicating Agents 1. Mission-Oriented Agents courtesy 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, 1996 Courtesy of Kanna Rajan, NASA Ames. Used with permission. 4

5

6 DTE Activity Name Night Time Rover Operations Pre-Comm Session Sequence Plan Durat on Current Sol Sequence Plan Review DTE period Sleep Current Sol Sequence Plan Review Night Time Rover Operations 2.00 Prior Sol Sequence Plan Review Prior Sol Sequence Plan Review 4.50 Real-TIme Monitoring Real-TIme Monitoring Real-TIme Monitoring 0.75 Downlink Product Generation Downlink Product Generation Tactical Science 5.00 Tactical Science Assessment/Observation Planning Assessment/Observation Planning 1.00 Science DL Assessment Meeting Science DL Assessment Meeting Payload DL/UL Handoffs 0.50 Payload DL/UL Handoffs Tactical End-of-Sol Engr. Assessm 5.50 Tactical End-of-Sol Engr. Assessment & Planning Planning 0.50 DL/UL Handover Meeting DL/UL Handover Meeting Skeleton Activity Plan Update 2.50 Skeleton Activity Plan Update SOWG Meeting 2.00 SOWG Meeting 0.25 Uplink Kickoff Meeting Uplink Kickoff Meeting 1.75 Activity Plan Integration & Validatio Activity Plan Integration & Validation Activity Plan Approval Meeting 0.50 Activity Plan Approval Meeting Build & Validate Sequences 2.25 Build & Validate Sequences 1.00 UL1/UL2 Handover UL1/UL2 Handover 2.50 Complete/Rework Sequences Complete/Rework Sequence 0.75 Margin 1 Margin 1 Command & Radiation Approval 0.50 Command & Radiati Margin Margin Radiation Radiation 7.00 MCT Team 4.00 Wakeup DFE One day in the life of a Mars rover rover operation (7 hr) tactical sci assess & obs planning (5 hr) tactical end-of -sol eng assess (5 hr) SOWG mtg(2 hr) activity integration & validation (3.5 hr) sequence development (5.5 hr) sequence integration & validation (4 hr) Courtesy of Kanna Rajan, NASA Ames. Used with permission. Agent Building Blocks (objective 2) Activity Planning Execution/Monitoring Types of Agents (objective 1) 1. Mission-oriented Agents 2. Self-repairing Agents 3. Mobile Agents 4. Agile Agents 5. Communicating Agents 6

7 7 year cruise ~ ground operators 2. Self-Repairing Agents Affordable Missions ~ 1 billion $ 150 million $ 7 years to build 2 year build 0 ground ops Cassini Maps Titan courtesy JPL Four launches in 7 months Mars Climate Orbiter: 12/11/98 Mars Polar Lander: 1/3/9 Stardust: 2/7/99 QuickSCAT: 6/19/98 courtesy of JP Mars Polar Lander Spacecraft require a good physical commonsense Launch: 1/3/99 courtesy of JPL 7

8

9 Remote Agent Goals Goal-directed First time correct projective reactive Commonsense models Mission Description Executive Scripts Heavily deductive Planner/ Scheduler Diagnosis & Repair Mission-level actions & resources component models Remote Agent Experiment May 17-18th experiment Generate plan for course correction and thrust Diagnose camera as stuck on Power constraints violated, abort current plan and replan Perform optical navigation Perform ion propulsion thrust May 21th experiment. Diagnose faulty device and Repair by issuing reset. Diagnose switch sensor failure. Determine harmless, and continue plan. Diagnose thruster stuck closed and Repair by switching to alternate method of thrusting. Back to back planning Agent Architecture (Objective 1) Plan Monitor & Diagnosis Execute 9

10 Agent Building Blocks (Objective 2) Activity Planning Execution/Monitoring Diagnosis Repair Scheduling Resource Allocation Types of Agents (Objective 1) 1. Mission-oriented Agents 2. Self-repairing Agents 3. Mobile Agents 4. Agile Agents 5. Communicating Agents 3. Mobile Agents Day 2 Initial Position; Followed by Close Approach Day 2 Traverse Estimated Error Circle Target Day 2 Traverse Estimated Error Circle Day 1 Long-Distance Traverse (<20-50 meters) During the Day Autonomous On-Board Navigation Changes, as needed Day 3 Science Prep (if Required) Day 4 During the Day Science Activities Courtesy of Kanna Rajan, NASA Ames. Used with permission. 10

11

12 Agent Building Blocks (Objective 2) Activity Planning Execution/Monitoring Diagnosis Repair Scheduling Resource Allocation Path Planning Localization Map Building Agent Architecture (Objective 1) Locate in World Monitor & Diagnosis Plan Execute Map Plan Routes 13

13 Types of Agents (objective 1) 1. Mission-oriented Agents 2. Self-repairing Agents 3. Mobile Agents 4. Agile Agents 5. Communicating Agents 4. Agile Agents 14

14 Agent Building Blocks Activity Planning Execution/Monitoring Diagnosis Repair Scheduling Resource Allocation Path Planning Localization Map Building Trajectory Design Policy Construction Agent Architecture (Objective 1) Locate in World Monitor & Diagnosis Plan Execute Map Plan Routes Maneuver and Track Types of Agents (objective 1) 1. Mission-oriented Agents 2. Self-repairing Agents 3. Mobile Agents 4. Agile Agents 5. Communicating Agents 15

15 5. Communicating Agents Agent Building Blocks (Objective 1) Activity Planning Execution/Monitoring Diagnosis Repair Scheduling Resource Allocation Path Planning Localization Map Building Trajectory Design Policy Construction Plan Adaptation Dialogue Management People Tracking Agent Architecture (Objective 1) Communicate and Interpret Locate in World Monitor & Diagnosis Plan Execute Map Plan Routes Maneuver and Track 16

16

17 Example: surface exploration ERA EVA robotic assistant follows astronaut and helps with sample collection, instrument placement Example Mission Scenario: Task Execution Robot walks to its sample area Begins collecting samples Walks back to astronaut Stumbles over unseen rock along the way, but recovers using appropriate limb motions 18