cmu 15 781
play

CMU 15-781 Lecture 21: Multi-Robot Systems Teacher: Gianni A. Di - PowerPoint PPT Presentation

Jan 07, 2024 •202 likes •521 views

CMU 15-781 Lecture 21: Multi-Robot Systems Teacher: Gianni A. Di Caro M ULTI -R OBOT S YSTEMS ? 15781 Fall 2016: Lecture 18 2 M ULTI -R OBOT S YSTEMS ? So far: How to represent world and knowledge How to make rational decisions How

  1. CMU 15-781 Lecture 21: Multi-Robot Systems Teacher: Gianni A. Di Caro

  2. M ULTI -R OBOT S YSTEMS ? 15781 Fall 2016: Lecture 18 2

  3. M ULTI -R OBOT S YSTEMS ? So far: How to represent world and knowledge • How to make rational decisions • How to learn to make rational decisions • How to take decisions as a collective • Our rational (AI) agent was quite abstract → Physical AI agents • Systems of multiple physical agents embedded in environments subject to the laws of physics • Subject to physical constraints and limitations for motion/action, perception, communication, computation • Partial knowledge and uncertainty are inherent • Autonomy in acting and decision-making 15781 Fall 2016: Lecture 18 3

  4. W HY “M ULTI ”-R OBOT S YSTEMS ? • Some tasks needs 2 or more robots • Linear / superlinear speedups • Parallel and spatially distributed system • Redundancy of resources ➔ Robustness • A robot ecology is being developed … • Environment inherently dynamic • Complex g-local interactions • Access shared resources • Need for (some) coordination • Increased (state) uncertainty • Communication issues • Costs / Benefits ratio • Practical problems × N 15781 Fall 2016: Lecture 18 4

  5. B ASIC T AXONOMY He Heteroge ogeneou ous sy syst stem em: Hom Homoge ogeneou ous sy syst stem em: different members have different skills members are interchangeable Loos oosely cou oupled: Tigh ghtly cou oupled: Being together is an advantage They need each other to successfully but not a strict necessity complete the team task Speedup Cooperation, Coordination 15781 Fall 2016: Lecture 18 5

  6. B ASIC T AXONOMY Coop ooperat ative (Benevol olent) : Robots are working together, forming a team Com ompetitive: Robots competing for resources, are in adversarial scenario 15781 Fall 2016: Lecture 18 6

  7. B ASIC T AXONOMY Central alized con ontrol ol Decentral alized/D /Distributed con ontrol ol 15781 Fall 2016: Lecture 18 7

  8. C ENTRAL P ROBLEM : M ULTI -R OBOT T ASK A LLOCATION (MRTA) Team Mission Decomposition in sub-tasks Team resources and status Who does what? (and when, how) Optimizing team Dependencies performance (tasks, agents) 15781 Fall 2016: Lecture 18 8

  9. MRTA: A F ORMAL D EFINITION (O PT ) Gi Given: ü A set of tasks, 𝑈 ü A set of robots, 𝑆 ü ℜ = 2 ' is the set of all possible robot sub-teams (e.g., ( 𝑠 ) = 0, 𝑠 , = 0, 𝑠 - = 1,𝑠 / = 0, 𝑠 0 = 1 ) ü A robot sub-team utility (or cost) function: 𝒱 𝑠 : 2 3 → ℝ ∪ { ∞ } (the utility/cost sub-team r incurs by handling a subset of tasks) ü An allocation is a function 𝐵: 𝑈 → ℜ mapping each task to a subset of robots. ℜ 3 is the set of all possible allocations Find: Fi Ø The allocation 𝐵 ∗ ∈ ℜ 3 that maximizes (minimizes) a global, team- level utility (objective) function 𝒱: ℜ 3 → ℝ ∪ { ∞ } 15781 Fall 2016: Lecture 18 9

  10. I NTENTIONAL / E MERGENT • Explicit/intentional TA: robots explicitly cooperate and tasks are explicitly assigned to the robot Batch/ Matching Online • Emergent TA: tasks are assigned as the result of local interactions among the robots and with the environment 15781 Fall 2016: Lecture 18 10

  11. T ASKS (Zlot, 2006) 15781 Fall 2016: Lecture 18 11

  12. U TILITY FUNCTION • Q and C are somehow estimat ates that account for all uncertainties, missing, information, … • Optimal allocation: Optimal based on all the available information → Rational decision-making • For some problems, an agent’s (sub-team’s) utility for performing a task is independent of of its utility for or perfor orming g an any ot other tas ask. • In general, this is not always true • Our definition fails capturing dependencies 15781 Fall 2016: Lecture 18 12

  13. B ASIC T AXONOMY ( Gerkey and Mataric , 2006) Assumption: Individual tasks can be assigned independently of each other and have independent robot utilities 15781 Fall 2016: Lecture 18 13

  14. W HY A T AXONOMY ? • A lot of “different MR scenarios” • A lot of “different” MRTA methods • Analysis and comparisons are difficult! • Taxonomy → Single out core features of a MRTA scenario • Allow to understand the complexity of different scenarios • Allow to compare and evaluate different approaches • A scenario is identified by a 3-vector (e.g., ST-MR-TA) 15781 Fall 2016: Lecture 18 14

  15. ST-SR-IA: L INEAR A SSIGNMENT If | R |=| T | the problem becomes a linear assignment and a polynomial-time solution exist! | R | | T | The Hungarian algorithm P P U rt x rt max has complexity O (| T | 3 ) r =1 t =1 | R | In a centralized P t = 1 , . . . | T | s.t. x rt = 1 architecture, with each r =1 robot sending its | T | | T | P r = 1 , . . . | R | x rt = 1 utilities to the t =1 controller, O (| T | 2 ) x rt ∈ { 0 , 1 } messages are needed Assignment with hundreds of robots in < 1s 15781 Fall 2016: Lecture 18 15

  16. ST-SR-IA: L INEAR A SSIGNMENT What if | R | ≠ | T | ? • • To preserve polynomial time solution, “ dummy ” robots or tasks can be included in a two-step process • If | R | < | T |: (| T |-| R |) dummy robots are added and given very low utility values with respect to all tasks, such that that their assignment will not affect the optimal assignment of | R | tasks to the “real” robots • The remaining | T |-| R | tasks (i.e., assigned to the dummy robots) can be optimally assigned in a second round, which will likely feature # of robots greater than the # of tasks • Dummy tasks with very low, flat, utilities are introduced such that their assignment will not affect the assignment of real tasks 15781 Fall 2016: Lecture 18 16

  17. ST-SR-IA: I TERATED A SSIGNMENT • Not always full/final task information is available since the beginning of the operations • How to deal with new / revised evidence (utility) in an iterative scheme? • Recompute from scratch or adapt greedily : Broadcast of Local Eligibility (BLE, 2001), worst-case 50% opt 15781 Fall 2016: Lecture 18 17

  18. ST-SR-IA: O NLINE A SSIGNMENT • Tasks are revealed one at-a-time • If robots can be reassigned , then solving each time the linear assignment provides the optimal solution MURDOCH (2002) When a new task is introduced, assign it to the most fit robot that is currently available. • Farthest Neighbor algorithm • Performance bound of FNA is the best possible for any on- line assignment algorithm (Kalyana-sundaram, Pruhs 1993). 15781 Fall 2016: Lecture 18 18

  19. ST-SR-TA: G ENERALIZED A SSIGNMENT | R | | T | P P U rt x rt max Robots gets a schedule of tasks r =1 t =1 | T | P r = 1 , . . . | R | s.t. c rt x rt ≤ T r t =1 | R | P t = 1 , . . . | T | x rt = 1 r =1 x rt ∈ { 0 , 1 } The “budget” constraints restricts the max number T r of tasks (or the total time/energy to execute them based on some cost parameter c ) that can be assigned to robot r NP-hard! 15781 Fall 2016: Lecture 18 19

  20. ST-SR-TA: G ENERALIZED A SSIGNMENT If dependencies / constraints are included, “more” NP-Hard → If the utility is related to traveling distances the problem falls in the class of m TSP, VRP problems Multi-robot routing 15781 Fall 2016: Lecture 18 20

  21. MT-SR-IA: G ENERALIZED A SSIGNMENT | R | | T | P P U rt x rt max Robots can work in || r =1 t =1 on multiple tasks | T | P r = 1 , . . . | R | s.t. c rt x rt ≤ T r t =1 | R | P t = 1 , . . . | T | x rt = 1 r =1 x rt ∈ { 0 , 1 } The “capacity” constraint explicitly restricts the max number T r of tasks that robot r can take, this time simultaneously Not common in the instances from MRTA NP-hard! 15781 Fall 2016: Lecture 18 21

  22. MT-SR-TA: VRP Robots can work in || on multiple tasks and have a time-extended schedule of tasks: quite uncommon in current MR literature Vehicle routing problems with capacity constraints and pick-up and delivery fall in this category: • Multiple vehicles transporting multiple items (goods, people) and picking up items along the way • Between a pick-up and delivery location the vehicle is dealing with MT • Visiting multiple locations is equivalent to TA NP-hard! 15781 Fall 2016: Lecture 18 22

  23. ST-MR-IA: S ET P ARTITIONING C OALITION F ORMATION • Model of the problem of dividing (partitioning) the set of robots into non-overlapping sub-teams ( coal oalition ons) to perform the given tasks instantaneously assigned • This problem is mathematically equivalent to set partitioning problem in combinatorial optimization. CT Cover (Partition) the elements in R x x x 1 (Robots) using the elements in CT x x 2 (feasible coalition-task pairs) R S x x 3 without duplicates (overlapping) x x 4 and at the min cost / max utility x x x 5 NP-hard! 15781 Fall 2016: Lecture 18 23

  24. MT-MR-IA: S ET C OVERING C OALITION F ORMATION • Model of the problem of dividing (partitioning) the set of robots into sub-teams ( coal oalition ons) to perform the given tasks instantaneously assigned. Overlap is admitted to model MT • This problem is mathematically equivalent to set covering problem in combinatorial optimization. CT Cover (Partition) the elements in R x x x (Robots) using the elements in CT 1 x x 2 (feasible coalition-task pairs) admitting R R x x 3 duplicates (overlapping) and at the x x 4 min cost / max utility x x x 5 NP-hard! 15781 Fall 2016: Lecture 18 24

Recommend

More recommend