Physical Path Planning using a Network of Learning Sensors: - - PowerPoint PPT Presentation

β–Ά
physical path planning using a network of learning
SMART_READER_LITE
LIVE PREVIEW

Physical Path Planning using a Network of Learning Sensors: - - PowerPoint PPT Presentation

Mar 16, 2024 558 likes β€’897 views

Physical Path Planning using a Network of Learning Sensors: Implementation and Testing Amir Rasouli The Active and Attentive Vision Lab ab Department of Electrical Engineering and Computer Science, York University, Toronto November 5, 2015

slide-1
SLIDE 1

Physical Path Planning using a Network of Learning Sensors: Implementation and Testing

The Active and Attentive Vision Lab ab Department of Electrical Engineering and Computer Science, York University, Toronto November 5, 2015

Amir Rasouli

slide-2
SLIDE 2

Swarm of Interacting Reinforcement Learners (SWIRLs)

  • A physical path planning approach
  • Network of sensors interacting to learn the

best path

  • Robots ask sensors where to go next

www.flickr.com

slide-3
SLIDE 3

SWIRLs Environments

slide-4
SLIDE 4

SWIRLs Stages of Operation

  • Initialization

π‘œ β†’ 𝑂(𝑦) 𝑅𝑦 π‘œ, π‘œ = 𝐹𝑑𝑒.π‘œ x n d …

𝐼𝐹𝑀𝑀𝑃(π‘¦π‘žπ‘π‘‘) 𝐼𝐹𝑀𝑀𝑃(π‘œπ‘žπ‘π‘‘) 𝐼𝐹𝑀𝑀𝑃(π‘’π‘žπ‘π‘‘) 𝐼𝐹𝑀𝑀𝑃(π‘œπ‘žπ‘π‘‘)

𝑅𝑦 𝑦, 𝑦 = 0 …

slide-5
SLIDE 5

SWIRLs Stages of Operation

  • Initialization
  • Estimation

x n d

πΉπ‘‡π‘ˆπ½π‘π΅π‘ˆπΉ( 𝑒𝑀[𝑦])

π‘…π‘œ

βˆ— 𝑒, 𝑒 + 𝑅𝑦 π‘œ, π‘œ β†’ 𝑅𝑦 π‘œ, 𝑒

π‘…π‘œ

βˆ— 𝑦, 𝑦 + 𝑅𝑒 π‘œ, π‘œ β†’ 𝑅𝑒(π‘œ, 𝑦)

𝑦 β†’ π‘Š 𝑒 𝑒 β†’ π‘Š 𝑦

πΉπ‘‡π‘ˆπ½π‘π΅π‘ˆπΉ( 𝑒𝑀[𝑒])

slide-6
SLIDE 6

SWIRLs Stages of Operation

  • Initialization
  • Estimation
  • Query

Ξ” 𝑅𝑉𝐹𝑆𝑍(𝑒) π‘ˆπ΅π‘†π»πΉπ‘ˆ(π‘œ) 𝑒𝑗𝑛𝑓𝑠 𝑒: 𝑑𝑒𝑏𝑠𝑒

x n d

slide-7
SLIDE 7

SWIRLs Stages of Operation

  • Initialization
  • Estimation
  • Query
  • Update

Ξ”

x n d

𝑒𝑗𝑛𝑓𝑠 𝑒: π‘‘π‘’π‘π‘ž π‘‰π‘„πΈπ΅π‘ˆπΉ(π‘œ, 𝑒)

slide-8
SLIDE 8

Implementations

  • Sequential Implementation
  • All processing and message passing is on a single thread
  • Semi-multi-threaded
  • Sensors and message passing on a single thread
  • Robots are run on individual threads
  • Multi-threaded
  • Sensors and robots are run on individual threads
  • Communication is through shared memory
slide-9
SLIDE 9

Code Structure

slide-10
SLIDE 10

Sequential Controller

package swirl; public class SwirlSequential { private final List<SensorSeq> sensorsList = new ArrayList<SensorSeq>(); private final List<RobotSeq> robotsList = new ArrayList<RobotSeq>(); … public void execute() { initialize(); List<Message> robotsRequestList = new ArrayList<Message>(); Map<Character, Message> targetList = new HashMap<Character,Message>(); … while(true) { if (timer(…)) { estimate();} for (int rIdx = 0; rIdx < this.robotsList.size(); rIdx++) { robotsRequestList.add(this.robotsList.get(rIdx).operate(…)); } … for (int m = 0 ; m < robotsRequestList.size(); m++) { … targetList.add(sensorsList.get(…).receive(robotsRequestList.get(m))); } } }

slide-11
SLIDE 11

Semi-Multi-Threaded Controller

package swirl; public class SwirlSemiMultiThreaded { private final ConcurrentMap<Character, Message> targetList = new ConcurrentHashMap<Character,Message>(); private final BlockingQueue<Message> robotsRequestList = new ArrayBlockingQueue<Message>(SwirlController.NUMBER_OF_ROBOTS*2); … public void execute() { … setThreads(); BlockingQueue<Message> rejectList = new ArrayBlockingQueue<Message>(SwirlController.NUMBER_OF_ROBOTS); while(true) { if (timer(…)){...} while (!robotsRequestList.isEmpty()) { … if (targetMsg.targetId == -1) { rejectList.offer(…); } … } while (!rejectList.isEmpty()) { this.robotsRequestList.offer(rejectList.poll()); } } } }

slide-12
SLIDE 12

Multi-Threaded Controller

package swirl; public class SwirlMultiThread { private final Map<Integer,BlockingQueue<Message>> initList = new HashMap<Integer,BlockingQueue<Message>>(); private final Map<Integer,BlockingQueue<Message>> estList = new HashMap<Integer,BlockingQueue<Message>>(); private final Map<Integer,BlockingQueue<Message>> queryList = new HashMap<Integer,BlockingQueue<Message>>(); private final Map<Integer,BlockingQueue<Message>> updateList = new HashMap<Integer,BlockingQueue<Message>>(); private final ConcurrentMap<Character, Message> targetList = new ConcurrentHashMap<Character,Message>(); … public void execute() { … setThreads(); … }

slide-13
SLIDE 13

Graphics

slide-14
SLIDE 14

Multi-Threading in Graphics2D

SwingWorker<Boolean, DrawSensor> worker = new SwingWorker <Boolean, DrawSensor>() { protected Boolean doInBackground() throws Exception { … DrawSensor s = new DrawSensor(…); publish(s); return true; } protected void process(java.util.List<DrawSensor> chunks) { frame.getContentPane().remove(…)); graphics.put((char)selfId,frame.getContentPane().add(chunks.get(chunks.size()-1))); frame.revalidate(); frame.repaint(); } }; worker.execute();

slide-15
SLIDE 15

Simulation

25 25 25

Sensor

Uninitialized Default Estimate

Robot

E = ID 44 = Destination

slide-16
SLIDE 16

Sequential Run

slide-17
SLIDE 17

Sequential Run

slide-18
SLIDE 18

Sequential Run

slide-19
SLIDE 19

Sequential Run

slide-20
SLIDE 20

Sequential Run

slide-21
SLIDE 21

Sequential Run

slide-22
SLIDE 22

Multi-Threaded Run

slide-23
SLIDE 23

Multi-Threaded Run

slide-24
SLIDE 24

Multi-Threaded Run

slide-25
SLIDE 25

Multi-Threaded Run

slide-26
SLIDE 26

Multi-Threaded Run

slide-27
SLIDE 27

Multi-Threaded Run

slide-28
SLIDE 28

Testing

  • Functionality
  • System behaves as expected
  • Fault-free
  • E.g. DeadLock, ConcurrentModification and NullPointer exceptions
  • Performance
  • What is the best concurrent structure
  • General Purpose Test
  • Unexpected issues
slide-29
SLIDE 29

Functionality

  • Case 1 : Sensors get to know all
  • ther sensors
  • Random experiments with 4-10

robots and 10-300 sensors

  • Case 2: Learning
  • Full learning hard to measure
  • Run the same experiments multiple

times

20 runs with identical configuration using 4 robots and 1280 sensors

slide-30
SLIDE 30

Fault-free

  • Deadlock caused by BlockingQueue
  • Null return from the queues
  • Concurrent modification of Maps
  • Ongoing detection by setting up all possible scenarios
slide-31
SLIDE 31

Performance

  • Case 1: Code Structure,
  • e.g. order of message

handling

  • Timing of estimates
  • Case 2: Handling Concurrency
  • ConcurrentMap is more

efficient than Collection.Synchronized()

  • Single-Channel vs Four-

Channel Communication no significant difference

Number of Robots Number of sensors 5 6 7 8 9 10 11 60 0.114

  • 11.307

0.031 0.29

  • 12.925

0.291 0.052 100 0.36

  • 4.494

1.035 0.473 1.105 0.894 0.346 140 1.277 2.064 1.866 1.383

  • 2.63

1.489 10.5 180 1.219 2.172 7.268 2.193 2.145

  • 9.714

6.359 220 2.088 4.8

  • 8.025
  • 6.699

2.815 1.325 9.11 260 7.564 1.671 12.176 2.793 9.035

  • 7.868

2.568 300 2.11 5.361 5.425 4.296 24.26 3.231 6.889

The results are obtained by subtracting single channel running times from the corresponding four channel trials

slide-32
SLIDE 32

Unexpected problems

  • Over 300 of experiments
  • Various systems
  • Intel Xeon 2.67 GHz with 24 single-threaded cores
  • Intel I7 2.00 GHz with 2 double-threaded cores
  • Intel Xeon 2.66 GHz with 8 single-threaded cores
  • Systematic change of parameters
  • Number of sensors
  • Number of robots
  • Maximum Speed of robots
  • Timing of estimate steps
slide-33
SLIDE 33

Future Work

  • Measure the performance of each implementation
  • Observe how each method scales to the size of the system