Bio-mimetic Robot Control Companion slides for the book Bio-Inspired - - PowerPoint PPT Presentation

▶
bio mimetic robot control
SMART_READER_LITE
LIVE PREVIEW

Bio-mimetic Robot Control Companion slides for the book Bio-Inspired - - PowerPoint PPT Presentation

May 17, 2023 571 likes •841 views

Bio-mimetic Robot Control Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, 1 Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press Mainstream A.I. A.I. was born in 1956 as a research field to

slide-1
SLIDE 1

1

Bio-mimetic Robot Control

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-2
SLIDE 2

2

Mainstream A.I.

  • A.I. was born in 1956 as a research field to replicate human intelligence
  • Over the years, most efforts concentrated on reasoning, planning, logic

Chess play was the hallmark of intelligence Until 1997: defeat of Kasparov by IBM DeepBlue

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-3
SLIDE 3

3

Machina Speculatrix (Grey Walter, 1953)

Reconstructed by Owen Holland

Simple analog electronics Simple phototaxis, obstacle avoidance Intelligence due to interaction with environment

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-4
SLIDE 4

Vehicles (Valentino Braitenberg, 1984)

Show behavioral effect of simple patterns of neural connectivity Intelligence is in the eye of the observer Attraction Avoidance Curiosity Memory Learning Intentionality … Artificial Evolution

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press 4

slide-5
SLIDE 5

5

WHY

Engineering motivation: Extract a principle from biology that solves a specific engineering problem that is not satisfactorily solved with

  • ther techniques

Scientific motivation: Use a robot to validate a model that is heavily based on physical embodiment and environmental situatedness

What is Bio-mimetic Control?

There is not a single methodology. Four case studies:

  • Vision-based flight (housefly)
  • Song recognition and localization (cricket)
  • Vision-based homing (ants, wasps)
  • Swim and walk (salamander)

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-6
SLIDE 6

6

INDOOR FLIGHT

Small size brings strong energy constraint, which rules out active sensors (active IR, sonar, laser) and computationally-expensive signal processing

Vision-Based Flyers

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-7
SLIDE 7

7

VISION-BASED INSECT FLIGHT

Flies and several other insects use modifications of optical image (Optic Flow) to avoid obstacles.

  • Distance to objects is proportional to OF magnitude
  • nly in straight flight.
  • Houseflies fly along straight trajectories interrupted

by rapid, saccade-like, rotations to avoid obstacle.

Insect Vision

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-8
SLIDE 8

8

VISION-BASED INSECT FLIGHT

Artificial Compound Eye

Circular array of analog Elementary Motion Detectors Analog electronics 1- Straight trajectory 2- Computation of obstacle distance map 3- New heading: goal direction and distance map

Franceschini, Pichon, Blanes, 1992 Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-9
SLIDE 9

9

30 g 15-minute autonomy 1.2 - 2.5 m/s

Indoor Flying Robot

Zufferey & Floreano, 2006

Two linear cameras pointing 45% from front

  • Front = small OF
  • Sides = large OF, but not useful for navigation

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-10
SLIDE 10

10

OFDiv = OFRight - OFLeft initiate rotation if OFDiv > threshold OFDiff = abs[OFRight] - abs[OFLeft] rotate towards OFDiff

When & Where to Turn

Zufferey & Floreano, 2006

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-11
SLIDE 11

11

CONTROL STRATEGY

Roll Stabilization

  • Halteres used to stabilize flight in housefly
  • MEMS piezo-electric gyroscopes in robot

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-12
SLIDE 12

12

slide-13
SLIDE 13

13

  • Female cricket recognizes male song and goes towards it
  • Song is composed of syllables of pure tone at 4.5 kHz
  • Syllable Repetition Interval is different for each species
  • Female goes only to crickets of same species

Cricket Song Recognition and Localization

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-14
SLIDE 14

14

OPEN ISSUES ON CRICKET’S BRAIN

How does female recognize song? How does it go towards song source? Is firing rate or firing time used by auditory neurons? Several models exist, but none is complete and well-specified

Existing Models

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-15
SLIDE 15

15 spike refractory period integration + leakage

x1 x2 x3 x4

Binary values

Spiking neuron

ARTIFICIAL NEURONS

Integrated Model with Spiking Neurons

  • Cricket advances at constant speed and turns when hears song
  • Synaptic strength decreased by controlateral AN, but exponential recovery
  • First AN to fire generates firing of ipsilateral MN
  • Too high Syllable Repetition Interval will neutralize both MN neurons
  • Too slow Syllable Repetition Interval will not generate correct trajectory

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-16
SLIDE 16

16

ROBOCRICKET, Edinburgh

Validation with Robot

Analog electronics model cricket auditory system Robot can track recorded songs from real crickets Robot ignores songs of other species

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-17
SLIDE 17

17

DESERT ANT NAVIGATION

No pheromone use because

  • f evaporation + scattered

food Strategies:

  • Path integration
  • Visual piloting
  • Systematic search

NEST PREY

Desert Ant Homing

slide-18
SLIDE 18

18

Insect may store retinotopic images of the visual scene and recover paths by aligning themselves so to match stored templates (Cartwright & Collett, 1983):

  • Assumes high memory capacity
  • Neural circuit may have to be complex

Snapshot Matching

Average Landmark Vector model (Lambrinos et al, 2000): One vector per landmark ALV: average of all landmark vectors Storage of target ALV Heading = ALV(cur) - ALV(tar)

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-19
SLIDE 19

19

VISUAL SNAPSHOTS

Sahabot

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-20
SLIDE 20

20

Homing by Image Navigation

Root Mean Square error of images form a surface with an increasing gradient as one moves away from home (target image) Insects may memorize home image at outward flight and compute RMS to sample gradient on return flight

Zeil et al. 2003

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-21
SLIDE 21

21

  • CPG is a set of neurons, that display rythmic oscillatory activity
  • When coupled with body, can be used to generate rythmic motion
  • Architecture, weight strengths, and time delays of connection affect behavior

Amphibot, Ijspeert EPFL

Rythmic Locomotion

CPG neurons oscillate within a frequency range Below or above that range, they stop working

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-22
SLIDE 22

22

Salamander’s behavior

  • Salamanders can both swim and walk in wave motion
  • Speed of waves are proportional to intensity of brain signals to CPG
  • Transition between walk and swim

low high fast swim swim transition fast walk walk still brain signal - behavior

Open questions:

  • One, two, or more CPG circuits?
  • Architecture?

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-23
SLIDE 23

23

Model of Salamander’s control circuit

MODEL ASSUMPTIONS (Ijspeert et al. 2007):

  • Body CPG + Limb CPG
  • Limb CPG oscillates at lower frequency and saturates at lower threshold
  • Connections from limb to body CPG are stronger than connections within body CPG

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-24
SLIDE 24

24

Transition between walk and swim

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press

slide-25
SLIDE 25

Bio-mimetic Conclusions

Can bring simplicity and efficiency to (some) control problems Behavioral complexity derives from embodied and situated control A bio-mimetic project needs clear definition of primary motivation:

  • engineering
  • scientific

Bio-mimetic approach is gaining adepts over classical control in robotics

Companion slides for the book Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies by Dario Floreano and Claudio Mattiussi, MIT Press 25