CS 599: Distributed Intelligence in Robotics

Transcription

1 CS 599: Distributed Intelligence in Robotics Winter Dr. Daisy Tang All lecture notes are adapted from Dr. Lynne Parker s lecture notes on Distributed Intelligence in Autonomous Robotics

2 Outline Overview of syllabus Overview of distributed intelligence in robotics

3 Introduction to Distributed Intelligence Distributed intelligence refers to systems of entities working together to reason, plan, solve problems, think abstractly, comprehend ideas and language, and learn.

4

5 Objective of Distributed Intelligence The objectiveis to generate systems of software agents, robots, sensors, computer systems, and even people and animals that can work together with the same level of efficiencyand expertise as human teams We will have a focus of study on multirobot system(mrs)

6 Rapidly Growing Research in MRS

7 How Rapidly is This Research Growing?

8 Reasons for the Rapid Growth Advances in individual autonomous robotics Advances in understanding of complex systems Increased computational capabilities Many potential application domains RoboCup influence Etc.

9 Potential Application Domains for MRS

10 Why Distributed Intelligence? Inherently distributed task High task complexity Efficiency through parallelism Robustness through redundancy Easier design of robots

11 Challenges How to manage the complete system? Lack of centralized control More communication requirement? Interference between entities Increased uncertainty about the system

12 MRS Design is Challenging Solutions are dependent upon task requirements and robot capabilities Efficiency Robustness Speed Flexibility Scalability Task 1 Task 2? Task 3

13 7 Primary Areas of Research in Distributed Robotics 1. Biological inspirations 2. Motion coordination 3. Communication 4. Object transport and manipulation 5. Reconfigurable robotics 6. Architectures, task planning, and control 7. Localization, mapping, and exploration

14 1. Biological Inspirations Objective: study biological systems to achieve engineering goals Communication Auditory, chemical, tactile, visual, electrical Direct, indirect, explicit, implicit Roles Strict division vs. loose assignments Hierarchies Purpose: reduction in fighting, efficiency Territoriality Reduces fighting, disperses group, simplifies interactions Leaf cutter ants Bee colony

15 Gist: Application of simple local control rules of various biological societies (ants, bees, and birds) to the development of similar behaviors in cooperative robot systems Flock, disperse, aggregate, forage, and follow trails Emergent cooperation as a result of acting on selfish interest

16 Swarm-Type Dispersion A. Howard, USC

17 Biological Inspirations: Future

18 2. Motion Coordination Issues studied: Multi-robot path planning Traffic control Formation Target tracking Target search Etc. Multi-robot path planning: enable robots to navigate collaboratively to achieve spatial positioning goals

19 Multi-Robot Motion Control: Keeping-Formation Work by Parker et ORNL Global control via local interactions

20 Motion Coordination: Future

21 3. Communication Objective: enable robots to exchange state and environmental information with a minimum bandwidth requirement Issues studied: Explicit vs. Implicit Local vs. Global Impact of bandwidth restrictions Variety of mediums: radio, IR, chemical scents, breadcrumbs, etc.

22 Communication: Future

23 4. Object Transportation and Manipulation Objective: enable multiple robots to collaboratively push, move, or carry objects that cannot be handled by one robot alone Issue studied: Two-robot teams vs. Swarm-type teams Cluttered vs. Uncluttered environments Global system models vs. Distributed models Etc.

24 Baton Passing and Box Pushing by Parker et al by Tang & Parker

25 Object Transport and Manipulation: Future

26 5. Reconfigurable Robotics Self-reconfiguring modular robots are autonomous kinematic machines with variable morphology Objective: obtain function from shape, allowing modules to reconnect to form shapes that achieve desired purpose Adapt to new circumstances Perform new tasks Recover from damage

27 Potential Applications Applications: Various navigation configurations (rolling track, spider, snake, etc.) Stair climbing, object support, etc. Space applications Requires long-term self-sustaining robots that can handle unforeseen situations and may require self repair

28 Movies of PolyBot By Mark Yim, Xerox PARC Porous material climbing Stair climbing Tricycle pedaling

29 Movies of CONRO (USC Information Science Institute) From snake to a rolling track Reconfiguration

30 Reconfigurable Robotic: Future

31 6. Architectures, Task Planning and Control Objective: development of overall control approach enabling robot teams to effectively accomplish given tasks Issues studied: Task allocation Action selection Delegation of authority and control Communication structure Heterogeneity vs. Homogeneity Achieving coherence amidst local actions Resolution of conflicts

32 Site-Clearing Movie by Tang & Parker

33 Architectures, Task Planning and Control: Future

34 7. Localization, Mapping and Exploration Objective: enable robot teams to cooperatively build models of their environment, or to accomplish spatial tasks requiring knowledge of other robot positions Issues studied: Extension of single-robot mapping approach to multi-robot teams Hardware, algorithms for robot positioning Sonar vs. Laser vs. Stereo vs. Sensor fusion Landmarks vs. Scan-matching

35 Monte Carlo Localization with 2 Robots by Thrun

36 3-D Mapping by Thrun The 3D map Was created with A laser pointing up At an angle

37 Localization, Mapping and Exploration: Future

38 Some New Research Areas Peer-to-peer human robot teams Highly heterogeneous teams Unmanned aerial and ground vehicles Physical demonstrations of large number of robots (> 100)

39 Topics We Will Study See our course website for details

40 Next Class Single robot control If you have taken Introduction to Robotics (CS 521) with me, you do not need to show up for the next class But you will need to have your player/stage installed and running