Behavior-based Multirobot Architectures. Why Behavior Based Control for Multi-Robot Teams? Multi-Robot control naturally grew out of single robot control.

Slides:



Advertisements
Similar presentations
ARCHITECTURES FOR ARTIFICIAL INTELLIGENCE SYSTEMS
Advertisements

Title: Intelligent Agents A uthor: Michael Woolridge Chapter 1 of Multiagent Systems by Weiss Speakers: Tibor Moldovan and Shabbir Syed CSCE976, April.
Some questions o What are the appropriate control philosophies for Complex Manufacturing systems? Why????Holonic Manufacturing system o Is Object -Oriented.
CLUSTERING IN WIRELESS SENSOR NETWORKS B Y K ALYAN S ASIDHAR.
The AGILO Autonomous Robot Soccer Team: Computational Principles, Experiences, and Perspectives Michael Beetz, Sebastian Buck, Robert Hanek, Thorsten Schmitt,
COORDINATION and NETWORKING of GROUPS OF MOBILE AUTONOMOUS AGENTS.
Introduction to Cyber Physical Systems Yuping Dong Sep. 21, 2009.
Embedded System Lab Kim Jong Hwi Chonbuk National University Introduction to Intelligent Robots.
COMMUNICATION IN MULTIROBOT TEAMS - BARATH CHRISTOPHER PETIT.
A Novel Cluster-based Routing Protocol with Extending Lifetime for Wireless Sensor Networks Slides by Alex Papadimitriou.
Lecture 6: Hybrid Robot Control Gal A. Kaminka Introduction to Robots and Multi-Robot Systems Agents in Physical and Virtual Environments.
Chapter 4 DECISION SUPPORT AND ARTIFICIAL INTELLIGENCE
Active SLAM in Structured Environments Cindy Leung, Shoudong Huang and Gamini Dissanayake Presented by: Arvind Pereira for the CS-599 – Sequential Decision.
Behavior-Based Formation Control for Multi-robot Teams Tucker Balch, and Ronald C. Arkin.
ECE 4340/7340 Exam #2 Review Winter Sensing and Perception CMUcam and image representation (RGB, YUV) Percept; logical sensors Logical redundancy.
Brent Dingle Marco A. Morales Texas A&M University, Spring 2002
Path Planning for Multi Agent Systems by Kemal Kaplan.
Experiences with an Architecture for Intelligent Reactive Agents By R. Peter Bonasso, R. James Firby, Erann Gat, David Kortenkamp, David P Miller, Marc.
A Free Market Architecture for Distributed Control of a Multirobot System The Robotics Institute Carnegie Mellon University M. Bernardine Dias Tony Stentz.
Autonomous Mobile Robots CPE 470/670 Lecture 8 Instructor: Monica Nicolescu.
Integrating POMDP and RL for a Two Layer Simulated Robot Architecture Presented by Alp Sardağ.
Motor Schema Based Navigation for a Mobile Robot: An Approach to Programming by Behavior Ronald C. Arkin Reviewed By: Chris Miles.
Location Systems for Ubiquitous Computing Jeffrey Hightower and Gaetano Borriello.
Humanoid Robotics – A Social Interaction CS 575 ::: Spring 2007 Guided By Prof. Baparao By Devangi Patel reprogrammable multifunctionalmovable self - contained.
Behavior- Based Approaches Behavior- Based Approaches.
Multiagent Systems: Local Decisions vs. Global Coherence Leen-Kiat Soh, Nobel Khandaker, Adam Eck Computer Science & Engineering University of Nebraska.
Measuring Cooperative Robotic Systems Using Simulation-Based Virtual Environment Xiaolin Hu Computer Science Department Georgia State University, Atlanta.
What is it? A mobile robotics system controls a manned or partially manned vehicle-car, submarine, space vehicle | Website for Students.
Introduction to Behavior- Based Robotics Based on the book Behavior- Based Robotics by Ronald C. Arkin.
Topic 2: Multi-Agent Systems a practical example categories of MAS examples definitions: agents and MAS conclusion.
Nuttapon Boonpinon Advisor Dr. Attawith Sudsang Department of Computer Engineering,Chulalongkorn University Pattern Formation for Heterogeneous.
DARPA Mobile Autonomous Robot SoftwareLeslie Pack Kaelbling; March Adaptive Intelligent Mobile Robotics Leslie Pack Kaelbling Artificial Intelligence.
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
September1 Managing robot Development using Agent based Technologies Dr. Reuven Granot Former Scientific Deputy Research & Technology Unit Directorate.
Cooperating AmigoBots Framework and Algorithms
Richard Patrick Samples Ph.D. Student, ECE Department 1.
NATIONAL TECHNICAL UNIVERSITY OF ATHENS DEPARTMENT OF MECHANICAL ENGINEERING CONTROL SYSTEMS LAB Georgios Rekleitis A Comparison of the Use of a Single.
Study on Genetic Network Programming (GNP) with Learning and Evolution Hirasawa laboratory, Artificial Intelligence section Information architecture field.
1 Robot Environment Interaction Environment perception provides information about the environment’s state, and it tends to increase the robot’s knowledge.
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
Synthetic Cognitive Agent Situational Awareness Components Sanford T. Freedman and Julie A. Adams Department of Electrical Engineering and Computer Science.
Multi-Robot Systems.
Chapter 40 Springer Handbook of Robotics, ©2008 Presented by:Shawn Kristek.
CS 415 – A.I. Slide Set 10. Controlling Multiple Robots Different considerations for multiple robots – Inherently dynamic environment – Complex local.
Mobile Robot Navigation Using Fuzzy logic Controller
Introduction to MultiRobot Systems. Lecture Outline Terminology Why group behavior is useful How group behavior can be controlled Why group behavior is.
Group Robotics. Last time we saw: Terminology Why group behavior is useful How group behavior can be controlled Why group behavior is very hard Approaches.
1 Distributed and Optimal Motion Planning for Multiple Mobile Robots Yi Guo and Lynne Parker Center for Engineering Science Advanced Research Computer.
DARPA ITO/MARS Project Update Vanderbilt University A Software Architecture and Tools for Autonomous Robots that Learn on Mission K. Kawamura, M. Wilkes,
Multi-Robot System Application
Distributed Algorithms for Multi-Robot Observation of Multiple Moving Targets Lynne E. Parker Autonomous Robots, 2002 Yousuf Ahmad Distributed Information.
Robotica Lecture Review Reactive control Complete control space Action selection The subsumption architecture –Vertical vs. horizontal decomposition.
Introduction to Artificial Intelligence CS 438 Spring 2008 Today –AIMA, Ch. 25 –Robotics Thursday –Robotics continued Home Work due next Tuesday –Ch. 13:
1 Multiagent Teamwork: Analyzing the Optimality and Complexity of Key Theories and Models David V. Pynadath and Milind Tambe Information Sciences Institute.
Selection of Behavioral Parameters: Integration of Case-Based Reasoning with Learning Momentum Brian Lee, Maxim Likhachev, and Ronald C. Arkin Mobile Robot.
University of Pennsylvania 1 GRASP Cooperative Control and Coordination of Multiple Robots Vijay Kumar GRASP Laboratory University of Pennsylvania
Learning for Physically Diverse Robot Teams Robot Teams - Chapter 7 CS8803 Autonomous Multi-Robot Systems 10/3/02.
Heterogeneous Teams of Modular Robots for Mapping and Exploration by Grabowski et. al.
University of Pennsylvania 1 GRASP Control of Multiple Autonomous Robot Systems Vijay Kumar Camillo Taylor Aveek Das Guilherme Pereira John Spletzer GRASP.
Marco Mamei Franco Zambonelli Letizia Leonardi ESAW '02
CS b659: Intelligent Robotics
Artificial Intelligence Lecture No. 5
Robot Teams Topics: Teamwork and Its Challenges
CIS 488/588 Bruce R. Maxim UM-Dearborn
Subsuption Architecture
Market-based Dynamic Task Allocation in Mobile Surveillance Systems
CHAPTER I. of EVOLUTIONARY ROBOTICS Stefano Nolfi and Dario Floreano
Learning to Communicate
Behavior Based Systems
Presentation transcript:

Behavior-based Multirobot Architectures

Why Behavior Based Control for Multi-Robot Teams? Multi-Robot control naturally grew out of single robot control Multi-Robot control naturally grew out of single robot control –Reactive: No state information –Planner: State space is already huge  Adding n additional robots to a state space of s results in an state space of s n –Hybrid: Same problems as a planning control system

Why Behavior Based Control for Multi-Robot Teams? Behavior Based Control Behavior Based Control Pros: Pros: –Since control is locally situated it scales well –No reliance on global communication or planning results in robots better able to handle sensor and actuator noise –Primitive Behaviors are relatively simple

Why Behavior Based Control for Multi-Robot Teams? Cons: Cons: –Difficult to create  Experimental –Difficult to analyze  Actions of robots depend on the actions of other robots  Behavior of the team is based on the interactions between robots instead of an individual robots control strategy

Issues in Behavior Based Multi- Robot Control How to create and combine behaviors to accomplish a given goal? How to create and combine behaviors to accomplish a given goal? How to coordinate robot behaviors? How to coordinate robot behaviors? –Use Communication? –What kind of knowledge should the team have?  Purely local control?  Hybrid local and global control?

Behavior Creation and Selection Bottom-Up Bottom-Up –Primitive behaviors should be minimalist in that sense that a primitive behaviors can not be derived from other primitive behaviors –Constrained by the robot’s physical capabilities –Constrained by the environment Top-Down Top-Down –Behaviors are constrained by the types of goals the must be accomplished by a team

Test Cases Equipment Equipment –20 mobile robots equipped with infra-red sensors, micro-switches, sonar, and radio Evaluation Evaluation –Repeatability –Stability –Robustness –Scalability

Test Cases Continued Primitive Behaviors Primitive Behaviors – Avoidance –Following –Aggregation –Dispersion –Homing –Wandering –Grasping / Dropping

Test Cases Continued Flocking Flocking –Summation of  Avoidance  Aggregation  Wandering –Addition of Homing for goal directed behavior Results Results –Goal directed behavior without dependence on a leader and robust in case of single robot failure –Flocking Flocking

Test Cases Continued Foraging Foraging –Temporally switch between: avoidance, dispersion, following, homing, and wandering Results Results –Basic behaviors were empirically shown to be robust and flexible in collecting pucks and dropping them off at a goal location

Reference Mataric, Issues and Approaches in the Design of Collective Autonomous Agents Mataric, Issues and Approaches in the Design of Collective Autonomous Agents

Behavior Based Multi-Robot Team Coordination Communication Communication –Often times relies on Master-Slave Hierarchy  Inherent brittleness to this approach –Bandwidth limitations –Robustness - Master failure? –Heterogeneous or Homogeneous approach? –Is explicit communication needed or is implicit communication enough to achieve the goal? –If communications are used how much is needed and what should be communicated?

Cooperation Without Communication Is cooperation in a behavior based multi- robot team without communication possible? Is cooperation in a behavior based multi- robot team without communication possible? If so, how effective is it? If so, how effective is it?

Behavioral Composition Forage Forage –Noise –Avoid static obstacles –Avoid robots Acquire Acquire –Move to goal –Avoid static obstacles –Noise Deliver Deliver –Move to goal –Avoid static obstacles –Avoid robots –Noise

Test Cases Simulation Simulation Map Size: 64 x 64 units Map Size: 64 x 64 units Maximum Sensor Distance: 25 units Maximum Sensor Distance: 25 units Forage Forage –Noise Gain: 1.2 –Noise Persistence: 4 –Avoid Obstacles Gain: 1.0 –Avoid Robots Gain: 0.5 Acquire/Deliver Acquire/Deliver –Noise Gain: 0.2 –Noise Persistence: 2 –Move to Goal Gain: 1.0 –Avoid Robots Gain: 0.1

Results 2 Robots / 1 Attractor 2 Robots / 1 Attractor

Results 4 robots / 4 attractors 4 robots / 4 attractors

Results Without using communication, the simulation still shows coherent cooperation between the team members Without using communication, the simulation still shows coherent cooperation between the team members –Cheaper hardware –Fewer points of failure

References Arkin, Cooperation without communication Arkin, Cooperation without communication For more quantitative comparisons between levels of communication see: Balch/Arkin, Communication in Reactive Multiagent Robotic Systems For more quantitative comparisons between levels of communication see: Balch/Arkin, Communication in Reactive Multiagent Robotic Systems

Behavior Based Multi-Robot Team Coordination Continued Local versus Global Control Laws Local versus Global Control Laws –Local Control  Simple and contain emergent properties  Oftentimes unclear as to how to design local control laws –Global Control  Allow for more coherent team cooperation  Often results in increased communication requirements

Global Control Laws Global Goal Knowledge Global Goal Knowledge –Information concerning the overall goal of the agents behavior –Can be encoded into robot if the goal is not dynamic Global Knowledge Global Knowledge –Information concerning what other robots are doing –Information concerning what other robots will do Obtaining this knowledge must often come from outside sources Obtaining this knowledge must often come from outside sources The knowledge is computationally costly The knowledge is computationally costly Oftentimes all the needed global knowledge is not known Oftentimes all the needed global knowledge is not known

Local Control Laws Computationally Simple Computationally Simple Handle dynamic environments well Handle dynamic environments well Oftentimes do not produce optimal results Oftentimes do not produce optimal results Must rely on physical sensors Must rely on physical sensors

Experiment Simulation of mission involving formation maintenance while moving to goal Simulation of mission involving formation maintenance while moving to goal 4 different strategies with increasing global control 4 different strategies with increasing global control Quantitatively measured via deviation from the formation and time taken to reach goal Quantitatively measured via deviation from the formation and time taken to reach goal

Experiment Continued Strategy I: Local Control Only Strategy I: Local Control Only –Effective for smooth trajectories –Sharp turns cause formation to break up due to local control  Robots maintain their position by staying a fixed distance from a certain side of neighbor

Experiment Continued Strategy II: Local Control Augmented by Global Goal Strategy II: Local Control Augmented by Global Goal –Robots given knowledge of global goal: Maintain line formation –Robot D now moves to a more globally appropriate position –May be inappropriate if B is merely avoiding obstacle

Experiment Continued Strategy III: Local Control Augmented with Global Goal and Partial Global Information Strategy III: Local Control Augmented with Global Goal and Partial Global Information –At time of robot B’s turn, other robots are informed of destination of waypoint X

Experiment Continued Strategy IV: Local control augmented by global goal and more complete global information Strategy IV: Local control augmented by global goal and more complete global information –Robots are given complete knowledge of leaders route –Allows other robots to predict future positions of the leader and resulting positions for themselves

Results As global information increases formation error and completion time decreases As global information increases formation error and completion time decreases Global goals are useful to incorporate if goals are known at run time Global goals are useful to incorporate if goals are known at run time Global information is useful in static, well defined environments Global information is useful in static, well defined environments

Results Continued Local control in situations where accomplishing the task as opposed to how the task accomplished often provides a suitable approximation of optimal behavior Local control in situations where accomplishing the task as opposed to how the task accomplished often provides a suitable approximation of optimal behavior Behavioral analysis in local control my approximate global knowledge Behavioral analysis in local control my approximate global knowledge General Rule: “Local control information should be used to ground global knowledge in the current situation. This allows the agents to remain focused on the overall goals of their group while reacting to the dynamics of their current situation.” General Rule: “Local control information should be used to ground global knowledge in the current situation. This allows the agents to remain focused on the overall goals of their group while reacting to the dynamics of their current situation.”

References Parker, Designing Control Laws for Cooperative Agent Teams Parker, Designing Control Laws for Cooperative Agent Teams

Questions ??? ???