robotics
play

Robotics 14 AI Slides (5e) c Lin Zuoquan@PKU 2003-2019 14 1 14 - PowerPoint PPT Presentation

Robotics 14 AI Slides (5e) c Lin Zuoquan@PKU 2003-2019 14 1 14 Robotics 14.1 Robots 14.2 Hardware 14.3 Perception 14.4 Motion planning 14.5 Controllor 14.6 Software AI Slides (5e) c Lin Zuoquan@PKU 2003-2019 14 2 Robots


  1. Robotics 14 AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 1

  2. 14 Robotics ∗ 14.1 Robots 14.2 Hardware 14.3 Perception 14.4 Motion planning 14.5 Controllor 14.6 Software AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 2

  3. Robots Robots are physical agents that perform tasks by manipulating the physical world Wide application: Industry, Agriculture, Transportation, Health, En- vironments, Exploration, Personal Services, Entertainment, Human augmentation and so on ⇐ Robotic age ⇒ Intelligent Robots AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 3

  4. Types of robots • Manipulators: physically anchored to their workplace e.g., factory assembly line, the International Space Station • Mobile robots: move about their environment – Unmanned ground vehicles (UGVs), e.g., The planetary Rover (in Mars), intelligent vehicles – Unmanned air vehicles (UAVs), i.e., drone – Autonomous underwater vehicles (AUVs) – Autonomous fight unit • Mobile manipulator: combined mobility with manipulation – Humanoid robots: mimic the human torso Other: prosthetic devices (e.g., artificial limbs), intelligent environ- ments (e.g., house equipped with sensors and effectors), multibody systems (swarms of small cooperating robots) AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 4

  5. Hardware A diverse set of robot hardware comes from interdisciplinary tech- nologies – Processors (controllors) – Sensors – Effectors – – Manipulators AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 5

  6. Sensors Passive sensors or active sensors – Range finders: sonar (land, underwater), laser range finder, radar (aircraft), tactile sensors, GPS – Imaging sensors: cameras (visual, infrared) – Proprioceptive sensors: shaft decoders (joints, wheels), inertial sen- sors, force sensors, torque sensors AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 6

  7. Manipulators P R R R R R Configuration of robot specified by 6 numbers 6 degrees of freedom (DOF) ⇒ 6 is the minimum number required to position end-effector arbitrarily. For dynamical systems, add velocity for each DOF. AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 7

  8. Non-holonomic robots θ ( x , y ) A car has more DOF (3) than controls (2), so is non-holonomic; cannot generally transition between two infinitesimally close configu- rations AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 8

  9. Perception Perception: the process mapping sensor measurements into internal representations of the environment – sensors: noisy – environment: partially observable, unpredictable, dynamic HMMs or DBNs can represent the transition and sensor models of a partially observable environment DNNs can recognize vision and various objects – the best internal representation is not known – unsupervised learning to learn sensor and motion models from data AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 9

  10. Perception generally Stimulus (percept) S , World W S = g ( W ) E.g., g = “graphics”. Can we do vision as inverse graphics? W = g − 1 ( S ) AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 10

  11. Perception generally Stimulus (percept) S , World W S = g ( W ) E.g., g = “graphics”. Can we do vision as inverse graphics? W = g − 1 ( S ) Problem: massive ambiguity! AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 11

  12. Perception generally Stimulus (percept) S , World W S = g ( W ) E.g., g = “graphics.” Can we do vision as inverse graphics? W = g − 1 ( S ) Problem: massive ambiguity! AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 12

  13. Perception generally Stimulus (percept) S , World W S = g ( W ) E.g., g = “graphics.” Can we do vision as inverse graphics? W = g − 1 ( S ) Problem: massive ambiguity! AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 13

  14. Localization Compute current location and orientation (pose) given observations (DBN) A t − 2 A t − 1 A t X t − 1 X t X t + 1 Z t − 1 Z t Z t + 1 AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 14

  15. Localization ω t Δ t x i , y i θ t + 1 v t Δ t h ( x t ) Z 1 Z 2 Z 3 Z 4 x t + 1 θ t x t Assume Gaussian noise in motion prediction, sensor range measure- ments AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 15

  16. Localization Can use particle filtering to produce approximate position estimate Robot position Robot position Robot position (b) (c) AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 16

  17. Localization Can also use extended Kalman filter for simple cases: robot landmark Assumes that landmarks are identifiable — otherwise, posterior is multimodal AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 17

  18. Mapping Localization: given map and observed landmarks, update pose distri- bution Mapping: given pose and observed landmarks, update map distribu- tion SLAM: given observed landmarks, update pose and map distribution Probabilistic formulation of SLAM (simultaneous localization and map- ping): add landmark locations L 1 , . . . , L k to the state vector, proceed as for localization AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 18

  19. Motion Planning Path planning: find a path from one configuration to another – various path planning algorithms in discrete spaces – path plan vs. task plan Continuous spaces: plan in configuration space defined by the robot’s DOFs conf-2 conf-1 conf-3 conf-3 conf-1 conf-2 e e s s Solution is a point trajectory in free C-space AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 19

  20. Configuration space planning Basic problem: ∞ d states! Convert to finite state space Cell decomposition: divide up space into simple cells each of which can be traversed “easily” (e.g., convex) become discrete graph search problem Skeletonization: reduce the free space to a one-dimensional representation identify finite number of easily connected points/lines that form a graph s.t. any two points are connected by a path on the graph AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 20

  21. Cell decomposition example 2DOFs robot arm workspace coordinates configuration space goal start goal start Problem: may be no path in pure freespace (white area) cells Solution: recursive decomposition of mixed (free+obstacle) cells AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 21

  22. Skeletonization: Voronoi diagram Voronoi diagram: locus of points equidistant from obstacles Problem: doesn’t scale well to higher dimensions AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 22

  23. Skeletonization: probabilistic roadmap A probabilistic roadmap is generated by random points in C-space and keeping those in freespace; create graph by joining pairs by straight lines Problem: need to generate enough points to ensure that every start/goal pair is connected through the graph AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 23

  24. Controllor Can view the motor control problem as a search problem in the dynamic rather than kinematic state space: – state space defined by x 1 , x 2 , . . . , ˙ x 1 , ˙ x 2 , . . . – continuous, high-dimensional Deterministic control: many problems are exactly solvable esp. if linear, low-dimensional, exactly known, observable Simple regulatory control laws are effective for specified motions Stochastic optimal control: very few problems exactly solvable approximate/adaptive methods ⇒ AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 24

  25. Biological motor control Motor control systems are characterized by massive redundancy Infinitely many trajectories achieve any given task E.g., 3-link arm moving in plane throwing at a target simple 12-parameter controller, one degree of freedom at target 11-dimensional continuous space of optimal controllers Idea: if the arm is noisy, only “one” optimal policy minimizes error at target I.e., noise-tolerance might explain actual motor behaviour AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 25

  26. Software Software architecture (of a robot car) SENSOR INTERFACE PERCEPTION PLANNING&CONTROL USER INTERFACE RDDF database Top level control Touch screen UI corridor pause/disable command Wireless E-Stop Laser 1 interface RDDF corridor (smoothed and original) driving mode Laser 2 interface Laser 3 interface road center Road finder Path planner laser map Laser 4 interface map trajectory VEHICLE Laser 5 interface Laser mapper INTERFACE vision map Camera interface Vision mapper Steering control obstacle list Radar interface Radar mapper Touareg interface vehicle state (pose, velocity) vehicle state Throttle/brake control GPS position UKF Pose estimation Power server interface vehicle state (pose, velocity) GPS compass IMU interface velocity limit Surface assessment Wheel velocity Brake/steering heart beats Linux processes start/stop emergency stop health status Process controller Health monitor power on/off data GLOBAL Data logger File system SERVICES Communication requests Communication channels clocks Inter-process communication (IPC) server Time server AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 26

  27. Pipeline architecture Execute multiple processes in parallel – sensor interface layer – perception layer – planning and control layer – vehicle interface layer AI Slides (5e) c � Lin Zuoquan@PKU 2003-2019 14 27

Recommend


More recommend