Multi-Agent Systems on Arduino & iOS

Slide 1

Slide 1 text

Tomáš Jukin @Inza Multi-Agent Systems on Arduino & iOS

Slide 2

Slide 2 text

Znáte Half-Life?

Slide 3

Slide 3 text

No content

Slide 4

Slide 4 text

No content

Slide 5

Slide 5 text

Table

Slide 6

Slide 6 text

No content

Slide 7

Slide 7 text

Robot

Slide 8

Slide 8 text

No content

Slide 9

Slide 9 text

No content

Slide 10

Slide 10 text

No content

Slide 11

Slide 11 text

Tomáš Jukin @Inza www.juicymo.cz @JuicymoCZ

Slide 12

Slide 12 text

Tomáš Jukin @Inza

Slide 13

Slide 13 text

Tomáš Jukin @Inza #MachineRoom at #DevFestCZ MachineRoom

Slide 14

Slide 14 text

No content

Slide 15

Slide 15 text

Movement Agent

Slide 16

Slide 16 text

Movement Agent Turning Agent

Slide 17

Slide 17 text

Movement Agent Turning Agent Border Agent

Slide 18

Slide 18 text

No content

Slide 19

Slide 19 text

When to use MAS? And for what?

Slide 20

Slide 20 text

When to use MAS? “When the solution is complex algorithm …”

Slide 21

Slide 21 text

When to use MAS? “… which is difﬁcult / impossible to design or code”

Slide 22

Slide 22 text

When to use MAS? Lets split complex behavior to simple agents

Slide 23

Slide 23 text

HOW to use MAS? A) MANY agents in ONE robot form population of ONE robot ! B) ONE agent per ONE robot form population of MANY robots

Slide 24

Slide 24 text

A) MANY agents per robot ! B) ONE agent per robot,   MANY robots HOW to use MAS?

Slide 25

Slide 25 text

A) MAS controls one robot ! B) MAS controls MANY robots HOW to use MAS?

Slide 26

Slide 26 text

Today I will talk about A) MANY agents in ONE robot form population of ONE robot MAS controls one robot HOW to use MAS?

Slide 27

Slide 27 text

When to use MAS? Agents do not need to know HOW to solve the problem

Slide 28

Slide 28 text

When to use MAS? Let them accumulate money instead ;) (for example)

Slide 29

Slide 29 text

When to use MAS? Let the emergence do the dirty work for us!

Slide 30

Slide 30 text

Any dangers? Emergence is not a cute girl…

Slide 31

Slide 31 text

To tune it properly can be tricky…

Slide 32

Slide 32 text

Examples? Airplane navigation Robot control Network diagnostics Predictive systems Simulations Computer Games AI

Slide 33

Slide 33 text

MASculptor Leden 2014 Objective-C Open Source

Slide 34

Slide 34 text

MASculptor example screenshots, few info iOS agnostic demo iOS6 only

Slide 35

Slide 35 text

Swift

Slide 36

Slide 36 text

What is an agent?

Slide 37

Slide 37 text

Co je to agent? Agent je entita zkonstruována za účelem kontinuálně a do jisté míry autonomně plnit své cíle v adekvátním prostředí na základě vnímání prostřednictvím senzorů a prováděním akcí prostřednictvím aktuátorů. Agent přitom ovlivňuje podmínky v prostředí tak, aby se přibližoval k plnění cílů. “ ”

Slide 38

Slide 38 text

Slide 39

Slide 39 text

Co je to agent? Agent Prostředí Senzory Aktuátory

Slide 40

Slide 40 text

Co je to agent? Agent Prostředí Senzory Aktuátory Vnitřní architektura Znalostní báze Řídíci jednotka

Slide 41

Slide 41 text

Rozdíl je ve vnitřní architektuře Agent Prostředí Senzory Aktuátory Vnitřní architektura

Slide 42

Slide 42 text

Naive (communicating) agent Reactive agent Deliberative (planning) agent Hybrid agent (reactive + planning)

Slide 43

Slide 43 text

performReasoning() { // ... } WorldState1 -> Action1 WorldState2 -> Action2 WorldState3 -> Action3 else -> idle Behavior Pattern Table Abstract Method Desire > Goal > Intents > Plan > Steps 1) Generate (choose) Plan 2) Execute Reactive + Deliberative WorldState1 -> Action1 WorldState2 -> Action2 WorldState3 -> Action3 else -> Deliberative 1) Generate (choose) Plan 2) Execute

Slide 44

Slide 44 text

What is an MAS?

Slide 45

Slide 45 text

What is MAS? Multi-Agent System is a system which uses group or population of agents interacting with environment in order to achieve global goal. Typically individual agents do not have a clue about the global goal. They have their own goals. The global goal is achieved by emergence.

Slide 46

Slide 46 text

Slide 47

Slide 47 text

No content

Slide 48

Slide 48 text

#Probee Robot

Slide 49

Slide 49 text

#Probee Robot 3x UltraSonic Sensor (Forward / Backward / Turret) 4x DC Motor 1x Motor Controller (Left / Right) 1x Arduino UNO (= brain) 1x Bluetooth Module 1x 2x16 I2C LCD (= status display)

Slide 50

Slide 50 text

MachineRoom

Slide 51

Slide 51 text

How our MAS in #Probee Robot works?

Slide 52

Slide 52 text

MAS in #Probee Robot Written in Wiring/Processing (C dialect) Running on 8bit processor 32KB program memory, 2KB RAM

Slide 53

Slide 53 text

Concurrency vs. Parallelism

Slide 54

Slide 54 text

Parallelism time →

Slide 55

Slide 55 text

Parallelism Agent 1 Agent 2 Agent 3 time →

Slide 56

Slide 56 text

Parallelism Agent 1 Agent 2 Agent 3 How much work can be done in parallel? ! Every agent… time →

Slide 57

Slide 57 text

Concurrency time →

Slide 58

Slide 58 text

Concurrency time → Agent 1 Agent 2 Agent 3

Slide 59

Slide 59 text

Concurrency time → Agent 1 Agent 2 Agent 3 How much work is actually computed in parallel? ! Only one (serialization)

Slide 60

Slide 60 text

MAS in #Probee Robot Written in Wiring/Processing (C dialect) Running on 8bit processor 32KB program memory, 2KB RAM

Slide 61

Slide 61 text

MAS in #Probee Robot Written in Wiring/Processing (C dialect) Running on 8bit processor 32KB program memory, 2KB RAM Parallelism 4

Slide 62

Slide 62 text

MAS in #Probee Robot Written in Wiring/Processing (C dialect) Running on 8bit processor 32KB program memory, 2KB RAM Parallelism 4 Concurrency 1 (serialization)

Slide 63

Slide 63 text

MAS in #Probee Robot Written in Wiring/Processing (C dialect) Running on 8bit processor 32KB program memory, 2KB RAM Parallelism 4 Concurrency 1 (serialization) Agents 4

Slide 64

Slide 64 text

No content

Slide 65

Slide 65 text

A1 Stopper Agent

Slide 66

Slide 66 text

A1 Stopper Agent A2 Communicating Agent

Slide 67

Slide 67 text

A1 Stopper Agent A2 Communicating Agent A3 Planner Agent

Slide 68

Slide 68 text

A1 Stopper Agent A2 Communicating Agent A3 Planner Agent A4 Movement Agent

Slide 69

Slide 69 text

Stop when collision would happen Connect with user (send sensor data, receive motor data) “Plan” the next action Control movement based on orders from A2/A3

Slide 70

Slide 70 text

A1 Stopper A2 Communicator A3 Planner A4 Mover Environment Sensors Actuators Sensors Actuators Sensors Actuators Sensors Actuators

Slide 71

Slide 71 text

A1 Stopper A2 Communicator A3 Planner A4 Mover Sensors Actuators Sensors Actuators Sensors Actuators Sensors Actuators Environment Blackboard Motors Bluetooth a1_vars a2_vars a4_vars Ultrasonic Sensors a3_vars

Slide 72

Slide 72 text

No content

Slide 73

Slide 73 text

No content

Slide 74

Slide 74 text

MAS run even on places you do not expect! M

Slide 75

Slide 75 text

Anyone can make Software, try Hardware as well! A

Slide 76

Slide 76 text

Seize the opportunity and make your MAS today! S

Slide 77

Slide 77 text

No content

Slide 78

Slide 78 text

Slide 79

Slide 79 text

A M

Slide 80

Slide 80 text

S A M

Slide 81

Slide 81 text

S A MAS run even on places you do not expect! M

Slide 82

Slide 82 text

S Anyone can make Software, try Hardware as well! A MAS run even on places you do not expect! M

Slide 83

Slide 83 text

Seize the opportunity and make your MAS today! S Anyone can make Software, try Hardware as well! A MAS run even on places you do not expect! M

Slide 84

Slide 84 text

Interested? Build or Implement YOUR own MAS! You can start right now… …with any language… …or on Arduino!