Agents Supporting Cooperative and Self Interested Human .ppt

1、Agents Supporting Cooperative and Self Interested Human Interactions in Open, Dynamic Environments,Katia P. Sycara School of Computer Science Carnegie Mellon University Pittsburgh, PA. 15213 http:/www.cs.cmu.edu/softagents,2,Talk Outline,Agents in Open Environments Agents Supporting Human Teams Info

2、rmation processing (memory intensive) Tasks Planning Tasks Agents Supporting Organizations E-commerce activities (negotiation, coalition formation, auctions) Forward to the Past: Agent-Based Web Services,3,Vision: Agents on the Web,A Wired/Wireless World populated with interoperating agents not just

3、 data,4,Overall Research Goal,Develop multiagent technology that allows agents (cooperative and self-interested) to coordinate autonomously and also assist individuals and human teams in environments that are: time stressed distributed uncertain open (information sources, communication links and age

4、nts dynamically appear and disappear) Team members (humans and agents) are distributed in terms of: time and space expertise,5,Reusable Environment for Task Structured Intelligent Networked Agents,Adaptive, self-organizing collection of Intelligent Agents infrastructure that interact with the humans

5、 and each other. integrate information management and decision support anticipate and satisfy human information processing and problem solving needs perform real-time synchronization of actions route and present the right information to the right person at the right time adapt to user, task and situ

6、ation Develop schemes for autonomous agent coordination Multi-agent discovery and interoperation Multi-agent adaptivity and learning,6,Open Environments,No predefined structure Agents leave and join the society dynamically Communication is not ensured all the time Information sources may appear and

7、disappear,7,Generic Tasks in Open Environments,Agents must be able to:discover each other. We distinguish the notion of agent location from the notion of agent functionality. Location is found through Agent Name Services (ANS) Functionality/capability is found through Middle Agents interact/transact

8、 with each other compose results of their reasoning monitor progress of delegated tasks,8,The RETSINA Multi-Agent Organization,distributed adaptive collections of information agents that coordinate to retrieve, filter and fuse information relevant to the user, task and situation, as well as anticipa

9、te users information needs.,9,RETSINA Single Agent Architecture,10,Some RETSINA Applications,Aiding Human Teams in joint mission planning (using ModSAF as a simulated battlefield) Agent-aided aircraft maintenance E-commerce in wholesale markets (agent-based auctions and negotiation) Agent-based Supp

10、ly Chain Management Robot teams for de-mining Team Rescue Scenario (NEO) Agent-based financial portfolio management Agent-based “on the move” collaboration on mobile devices,11,Visualization of Agent Interactions,12,Agent Discovery and Interoperation,Discovery necessary in open environments Interope

11、ration necessary for heterogeneous agents Agents advertise their expertise/capabilities to middle agents Requester agents ask middle agents for agents with particular capabilities Middle agents match requests to advertisements and return results Communication protocols include formal semantics and o

12、ntologies for interoperation The discovery scheme enables system robustness through functional substitutability of agents Sycara, K., Klusch, M. Widoff, S. and Lu, J. “LARKS: Dynamic Matchmaking among Heterogeneous Agents in Cyberspace“, JAAMAS, vol 5, no. 2, July 2002.,13,Types of Interactions,Prov

13、iders and requesters interact with each other directly a negotiation phase to find out service parameters and preferences (if not taken into account in the locating phase) delegation of service Providers and requesters interact through middle agents middle agent finds provider and delegates hybrid p

14、rotocols Reasons for interacting through middle agents privacy issues (anonymization of requesters and providers) trust issues (enforcement of honesty; not necessarily keep anonymity of principals); e.g. NetBill,14,Broadcaster,Broadcaster,Requester,Provider 1,Provider n,Request for service,Broadcast

15、 service request,Delegation of service,Results of service request,Offer of service,15,Matchmaker,Matchmaker,Requester,Provider 1,Provider n,Request for service,Contact information of providers that match the request,Advertisement of capabilities +para.,Delegation of service,Results of service reques

16、t,16,Broker,Broker,Requester,Provider 1,Provider n,Delegation of service + preferences,Advertisement of capabilities + para.,Delegation of service,Results of service,Results of service,17,Contract Net,Manager,Requester,Provider 2,Provider n,Request for service + preferences,Offer of service,Delegati

17、on of service,Results of service,Offer of service,Provider 1,Broadcast,Broadcast,Offer of service,Results of Service,18,Performance of Match-made System,19,Performance of Brokered System,20,Hybrid Human-Agent Teams,Human and software agents working together as a team to perform complex tasks in a di

18、stributed environmentAgents providing information access as well as user-centered problem-solving and decision supportAgents monitoring team activity and the environment so that effective assistance can be provided,21,Human-Agent Teams,Agent Rolessupport for individual team memberssimple reactive ag

19、ents: manage and present information meaningfully, react to event stimuliplanning agents: present courses of action based on emerging eventssupport for team activitysituation assessment: provide information to the team on environmentfacilitate communication within the teamsupportive behaviours: corr

20、ecting other team member, requesting backupas an autonomous team membercannot use human team member roles directlyprobably feasible for information access, event monitoring, planning of member roles,22,Agents in Teams: Expected Improvements,Reduce time for human teams to arrive at a decision Allow t

21、eams to consider a broader range of alternatives Enable teams to flexibly manage contingencies (replan, repair) Reduce individual and team errors Increase overall team performance,23,NAWCTSD TeamWork Dimensions,24,Aiding & Cognitive Resources,We might improve team performance by: Making individual t

22、asks easier freeing cognitive resources for team coordination tasks Aiding aspects of individual task exercised in coordination activities Supporting team coordination tasks directly,25,TANDEM Synthetic Radar Task,Lab Simulation : moderate fidelity Aegis-based simulation Characteristics : Real-time,

23、 reactive & inflexible Task : Forced Pace, High Workload, Highly Dependent on Cooperation, Shared Information, Individual Action Cognitive Demands: High working memory load Subjects must access from menus or obtain from teammates five parameter values and their classifications in order to reach each

24、 of their individual targeting decisions Studies : contrasted agent aiding for reducing memory load with assistance in communication and cooperation,26,Tandem Experiments,Three team members (Alpha, Bravo, & Charlie) each responsible for a different decision (type, classify, intent) Each team member

25、has 3 menus each accessing 3 parameters Each team member has 3 pieces of data for his task, but the remaining two items must be obtained from teammates,27,Speed: 250 knots (Its an aircraft)Ini Altitude: 0 Feet Signal: Medium (Its surface,Climb rate: 300 ft/sec (air craft),Comm Time: 10 sec (air craf

26、t),User may need information from teammates,28,Agent Aiding Strategies,Supports Individuals,Task,Supports Team Work,Registry,Shows who has what,data,Facilitates coordination,Persistent Memory,Information Push,Accumulates values for,own task,Pushes accessed values to teammates,Reduces verbal,communic

27、ation,Reduces communication,errors,Preserves accessed values for own decision,Preserves accessed values for communication to team,29,Experimental Design,Between subject design with 4 conditions: Individual Memory agent Team Registry agent Team Push agent Control (no agent)Each task is defined by 5 p

28、arameter values, 3 of which a team member can access from menus, the other 2 are gotten from team mates Three team mates Alpha, Bravo, Charlie, each responsible for a decision (type, intent, classification),30,Experimental Design (cont),10 teams of 3 subjects in each condition (120 subjects) Each se

29、ssion contained 3 trials, 15 minutes each Each trial included 75 targets with 3 levels of target difficulty Target difficulty : hard (25 targets), medium (25 targets) & easy (25 targets),31,Individual Agent,*,*,*,*,*,*,*,*,*,*,*,*,*,Time : 00:14:25,Agent Window -TYPE- Speed: 27 Climb/Dive : -366 Sig

30、nal -CLASS- Bearing: Origin: Red_Sea Range: 1.4-INTENT- Countermeasures:None Electronic Warfare:Missile Lock : Clean,Hooked Target : 35,Radius : 50 nm,OPER A B C,000,270,180,090,*,*,*,*,*,*,*,*,*,*,*,Individual Memory,T SCORE: 1200 I SCORE : 1950,32,Team Clipboard Agent,*,*,*,*,*,*,*,*,*,*,*,Time :

31、00:10:25,-TYPE-Speed: 120 Climb/Dive: 0 Alt/Depth: Sig Strength: Medium Comm Time:,Hooked Target : 45,Radius : 50 nm,OPER A B C,000,270,180,090,*,*,*,*,*,*,*,*,*,*,*,Team Push for Alpha,T SCORE: 2500 I SCORE : 2800,33,Team Checklist Agent,*,*,*,*,*,*,*,*,*,Time : 00:09:25,Hooked Target : 23,Radius :

32、 50 nm,OPER A B C,000,270,180,090,*,*,*,*,*,*,*,*,*,*,*,Registry Agent,T SCORE: 2500 I SCORE : 2800,-TYPE-A B C Speed * B Alt/DepthA B Climb/DiveA B Signal StrengthB C Comm Time -CLASS-B Intel *A B BearingA C RangeB C Maneuver -INTENT-A C CountermeasuresA Electronic WarA B Missile Lock* C ResponseB

33、C Threat,34,Identification of Hard Targets,Copyright Katia Sycara 2002,35,Aiding Teams Helps more than Aiding Individuals for Hard Targets,Team Registry,Team Push,Individual Memory,Control,Hard Targets Correct,230,220,210,200,190,180,Copyright Katia Sycara 2002,36,MokSAF Collaborative Planning Task,

34、Lab Simulation : MokSAF lightweight agent-based planning environment using ModSAF terrain database and Retsina-like planner Characteristics : Deliberative, iterative & multiattribute Task : Self-Paced, Complex, Highly Dependent on Cooperation, Shared Information, Team Action Cognitive Demands: Compl

35、ex problem-solving, requires multi-attribute negotiation among subjects Studies : Comparisons between autonomous, cooperative, and critiquing route planning agents Payne, T., Sycara, K. and Lewis, M. “Varying the User Interaction within Multi-Agent Systems” , In Proceedings of the Fourth Internation

36、al Conference on Autonomous Agents, June 3-7, Barcelona, Spain, 2000. pp 412-418,37,Humans & Agents,Agents: have access to digital information in the infosphere cannot consider intangible objectives which are not part of that digital infosphere Humans: Understand Idiosyncratic and situation-specific

37、 factors local politics, non-quantified information, complex or vaguely specified mission objectives Dynamically changing situations Information, obstacles, enemy actions Problem: To share and combine human and agent information and resources,38,Soil,Rendezvous Point,River,Forest,Road,Building,Teamm

38、ates route,Freeway,Commanders route,Start Point,Constraint,MokSAF Display,39,Experiments,Map planning environment Teams of three subjects Three conditions Control (route critic) Agent Autonomous Planning Agent Cooperative Planning Agent Capability to express intangible constraints via physical artif

39、acts on the map,40,Planning Routes,41,MokSAF: Autonomous Agent with user supplied constraints,42,Cooperative Agent/hilighter mode,43,Sharing Plans,Subjects create individual routes to rendezvous point by drawing them asking agent to draw them When ready, subjects can share plans with other commander

40、s all routes will appear on screen Can communicate with each other via typing into a comm program messages go to one commander or all commanders categorized by subject,44,Mission Objectives (Performance Measures),All platoons arrive at the specified rendezvous point within a some agreed time frame C

41、reate an optimal route in terms of path length The route should not violate any physical constraints The route should not violate any social constraints (e.g., avoid this area because the roads are under construction) The route should pass through areas designated as “go-bys” Minimize sharing paths

42、with other teammates The team should take the total number and types of units specified by the mission briefing. Too few units is worse than too many units. An exact match is best.,45,Path Length,Route Times,Path Length, Route Times, and Fuel Usage were uniformly better for Aided Teams,46,Results Ve

43、hicle Selection & Successful Rendezvous,On the more difficult Session 2 Rendezvous: Teams using the Cooperative RPA most closely approximated reference performance Teams using the Autonomous RPA made slightly less appropriate decisions Teams using the Route Critic Control performed poorly sometimes

44、failing to rendezvous For the less difficult Session 3 Rendezvous: Performance retains ordering although differences are not significant,47,Errors in Vehicle Choice session 2,Cooperative,Autonomous,Control,Errors,13,12,11,10,9,8,7,6,Shuttle Launch,Several distributed range operators must collaborate

45、 to achieve a successful launch within the launch window or abort the mission in minimal timeResponsible for monitoring a particular area in the launch zoneNegotiate with other range operatorsMonitoring of several conditions, such as There should be no civilian or military vehicles in the path of th

46、e shuttle, in case of falling debrisThe weather conditions need to be such that the exhaust plumage does not fall on inhabited areas,Copyright Katia Sycara 2002,Shuttle Launch,Work environment isdistributed time-criticalinformation-richcommunication-intensiveIncreasingly, bottleneck on team performa

47、nce is not availability of information, but limits on human capabilities: perception, cognition, attention,Copyright Katia Sycara 2002,Supporting Human-Agent Teams in Shuttle Mission Launch,Copyright Katia Sycara 2002,Approach,Develop task models appropriate to the distributed workflow Develop cogni

48、tive models of key team members Develop software agents to support the team members and the team Evaluate the approach and resulting system,Copyright Katia Sycara 2002,Evaluation,Verification of task and cognitive models with human performance data Evaluate effectiveness of software agents using models and then through empirical testing in the laboratory and field settings Develop evaluation metrics to assess team performance,Copyright Katia Sycara 2002,Range Operations & Space Launch Safety,