Top Banner
1 Adaptive Robotics Adaptive Robotics COM2110 COM2110 Autumn Semester 2008 Autumn Semester 2008 Lecturer: Amanda Sharkey Lecturer: Amanda Sharkey
77

Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

Dec 30, 2015

Download

Documents

Lee Hicks

Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey. Robots in the news. 9:13am UK, Friday November 14, 2008 “British scientists have come up with the first robot that can mimic a person's expressions simply by watching their face.” - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

1

Adaptive RoboticsAdaptive RoboticsCOM2110COM2110

Autumn Semester 2008Autumn Semester 2008Lecturer: Amanda SharkeyLecturer: Amanda Sharkey

Page 2: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

2

Robots in the news

9:13am UK, Friday November 14, 2008 “British scientists have come up with the first robot that can

mimic a person's expressions simply by watching their face.” A disembodied humanoid robotic head with video camera eyes

and 34 internal motors under “Flubber” skin Works with 10 human emotions (happiness, sadness, concern) Can mimic human expressions that it recognises. Bristol Robotics Lab (Chris Melhuish) at UWE Robot head commissioned from David Hanson Can scan and “understand” images at 25 frames per second.

Page 3: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

3

Lect 1: what is a robot? Brief history of robotics Early robots, Shakey and GOFAI, Behaviour-based robotics Mechanisms and robot control (and biological inspiration) Lect 2: Grey Walter, Brooks and Subsumption Architecture. Lect 3: Adaptation and learning Lect 4: Artificial Neural Nets and Learning Lect 5: Evolutionary Robotics Lect 6: Swarm Robotics (reasons for, biological inspiration, local

control and communication, self-organisation and emergence).

Page 4: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

4

Robots as Biological Models Robots can be used to investigate

biological questions And to test hypotheses

Page 5: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

5

Biorobotic Modelling What is Biorobotics? The intersection of biology and robotics Common ground between robots and

animals: Both are moving, behaving, systems Both have sensors and actuators Both require an autonomous control system

to carry out tasks in a dynamic world.

Page 6: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

6

To count as biorobotic modelling a system should Be robotic (physically instantiated) Be biological (address a biological

hypothesis, or demonstrate understanding of a biological system). (Webb, 2001)

Page 7: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

7

Biorobotics:

“understanding biology to build robots and building robots to understand biology” (Webb, 2001)

Page 8: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

8

Synthetic modelling Understanding by building

Synthetic vs analytic Analytic approach in empirical sciences –

performing experiments on existing systems (human, ant, brain region) and analysing results

Synthetic: Idea of creating an artificial system that reproduces certain aspects of natural system

Page 9: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

9

Robots have (arguably) replaced computers as the metaphor to develop and test models of biological intelligence

Page 10: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

10

Early example – “electric dog” (Seleno) Loeb and animal taxis or forced movement Hammond and Miessner,(1912) “electric

dog” Loeb saw the “artificial heliotrope

machine” as confirmation of his theories

Page 11: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

11

Advantages of biorobotic models Robotic implementations force scientists to be

very concrete in specifying the design of biological system

Robot operating in the real world constrains the choice of models

Can help to produce testable hypotheses Robot models allow scientists to study interaction

of model with the environment Simulations may introduce misleading

simplifications

Page 12: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

12

Webb (2001) 7 dimensions along which robotic models of biology differ1. Relevance: whether model tests and generates hypotheses

applicable to biology2. Level: the elemental units (atoms – society)3. Generality: the range of biological systems the model can represent4. Abstraction: the complexity relative to the original5. Structural accuracy: how well the model represents the actual

mechanisms underlying behaviour6. Performance match: to what extent the model behaviour matches

the target behaviour7. Medium: the physical basis by which the model is implemented.

Page 13: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

13

A robotic cricket: Song Recognition and Localisation Lund, Webb and Hallam (1998) Phonotaxis in the cricket Lund, H.H., Webb, B., and Hallam, J. (1998) Physical and

Temporal Scaling Considerations in a Robot Model of Cricket Calling Song Preference. Artificial Life, 4, 1, 95-107.

Webb, B., and Scutt, T. (2000) A simple latency-dependent spiking-neuron model of cricket phonotaxis. Biological Cybernetics, 82: 247-269.

Page 14: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

14

Female crickets recognise males’ songs and approach them.

They have an ear in each foreleg that produces direction-dependent differences in response amplitude

Turns to strongest response? But – only approach males of the same

species with characteristic song.

Page 15: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

15

Field cricket

Page 16: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

16

Ground cricket

Page 17: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

17

Tree cricket

Page 18: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

18Mole cricket

Page 19: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

19Bush cricket

Page 20: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

20Scaly cricket

Page 21: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

21

Taxis, or approach to a sensory source When modelled, often abstracted as a source

signal, whose value decreases with distance, and an animat that can sense that value and use it to control movement to the source.

E.g Braitenberg’s vehicle 2 But does this capture the essence of the

behaviour? Lund et al (1998) argue it does NOT

Page 22: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

22

They say … Abstract model ignores the real physics of signal

propagation and detection Reasonable approximation of light but not of

other modalities E.g. chemotaxis, where odour signal is highly

dispersed and carried by currents in environment Phonotaxis: sensors don’t respond exclusively to

relevant signal

Emphasis on physical properties of task and agent

Page 23: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

23

Unidirectional communication- Male cricket stridulates

- Rubbing wings together, or a leg against a wing.

- Female cricket responds by moving towards the source of the song.

Page 24: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

24

Page 25: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

25

Earlier implementation: LEGO robot – but slow and comparison to cricket data difficult.

Here, new robot and revised models band-pass selectivity using signals that are temporally identical to

crickets.

Page 26: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

26

The female cricket has 4 auditory openings An ear (tympanum) on each foreleg An auditory spinacle (hole) on each side of

the frontal section The 4 openings are linked by tracheal tubes

Sound reaches tympani through air And via internal tubes from other auditory

openings

Page 27: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

27

Page 28: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

28

Sounds transduced from tympani are a combination of delayed and filtered signals

The delays and filters improve cricket’s ability to discriminate the arrival direction of its mate’s song.

Sounds arriving from same side as tympanum are delayed (by internal structure) to arrive in antiphase to ipsilateral ear, and in phase with contralateral ear.

Sounds subtracted, so intensity at ipsilateral (same side) ear is enhanced

Page 29: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

29

Female crickets home in on potential partners by listening to their chirping (phonotaxis)

Sound is made of short bursts or syllables.

She moves towards the source of the song;

But she only responds to calling songs with syllable rates within a certain bandpass.

Are two processes implicated?

(i) recognition of cricket song

(ii) localisation system to approach source of the song.

Page 30: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

30

Page 31: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

31

Page 32: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

32

Page 33: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

33

The neural model Input from auditory sensors is fed into a neuron on

each side (N1) Activation can flow from N1 to another neuron N2

that feeds activation to the motor on that side. Activation modelled with a leaky integrator with

associated threshold (T high) before firing. When it is fired, activation has to decay below a

lower threshold (T low) before it can fire again NB N1 and N2 are not intended to correspond to

specific neurons in the cricket, but represent processes carried out by small numbers of neurons.

Page 34: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

34

Page 35: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

35

Phonotaxis experimental results

The robot is attracted to the male cricket calling song. It also discriminates between calling songs with the right carrier frequency (4.7kHz) and those with other carrier frequences (e.g. 6.7kHz)

Neural circuitry is set up to respond to species-specific syllable repetition interval

Page 36: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

36

Page 37: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

37

Page 38: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

38

The robot’s response to different syllable rates (little response when syllable intervals are less than 20 ms and more than 60 ms) is the same as in the real cricket.

Page 39: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

39

Their conclusions Explains why crickets respond only to certain calls Mechanism implements the bandpass filtering found in

crickets. With N1 neurons acting as low pass filter, and N2 acting

as high pass filter. “high level” model of cricket controller – N1 and N2

don’t correspond to specific neurons in cricket, but represent processes that may be carried out by 3-10 neurons in the cricket prothoracic ganglion and brain.

Page 40: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

40

Mechanism underlying phonotaxis in robotic cricket is surprisingly simple

Shows that there is not need for two neural control systems One to recognise the call One to locate it.

Here the two systems are one

Page 41: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

41

“A generic simulation may tell us little about real problems in approach behaviour. By investigating a specific biological system and modelling it at a level of detail driven by biological questions, we gain more sophisticated insights into the real problems of sensorimotor control” Lund et al (1998)

Page 42: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

42

But it is also possible to investigate phonotaxis in real crickets …..

Page 43: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

43

Hedwig, B., and Poulet, J. (2004) in Nature Using tracking ball, could look at cricket

responses to chirps presented from left and right of cricket

Response was very rapid - “reflex like reactive steering towards individual

sound pulses” Other methods (bandpass filtering brain neurons,

or template matching), would require at least two consecutive sound pulses

But recognition process may modulate steering responses over time.

Page 44: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

44

Biorobotics and …. Biological relevance: biorobotics more likely to

assume a particular biological target than biologically inspired robotics

Level: usually sensor transductions considered Generality: usually biorobotic models are more

specific e.g modelling sound localisation of crickets. But general principles can emerge (I.e. idea of exploiting timing properties of neural firing)

Page 45: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

45

Abstraction. Usually modelling simplifies some aspects – e.g. in the robot cricket model there is minimal representation of biological details in motor control of cricket

Structural accuracy: how well do mechanisms reflect real mechanisms in target? How would we know?

Usual aim is to build complete model that connects sensing to acting.

Page 46: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

46

Performance Match: to what extent does model behave like the target?How will model’s performance be assessed?Little possibility of falsifying model- if match is poor this could be attributed to measurements, not model- depends on interpretation of model behaviour

- parameters of model can be tuned until they match.

Page 47: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

47

Medium: what is simulation built from? Biorobotics, emphasis on “complete model” that

produces an output, and on understanding the environmental conditions

Page 48: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

48

What does a biorobotic model tell us? Can we ever really validate a model? Do models ever tell us something we didn’t

already know?

Page 49: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

49

Modelling: involves the correspondence between a real target system and something else.

Problems: underdetermination. There are many possible mechanisms that could underlie a behaviour. If two systems behave the same, it does not follow that the cause or mechanism of the behaviour is the same.

But, can show that a proposed mechanism is sufficient to produce the behaviour

Page 50: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

50

Can we assume that this model of cricket phonotaxis represents the mechanism underlying the cricket’s behaviour?

Reasons for doubting:

Page 51: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

51

Cataglyphus – the desert ant

Page 52: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

52

How does the desert ant return to nest after foraging? Cannot use pheromones – they evaporate too

quickly It uses a combination of strategies

Path integration Visual piloting Systematic search

Page 53: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

53

Main strategy: path integration Uses compass information based on

polarization pattern of the sky Directions of polarization (e-vectors) form

regular patterns, as sunlight is scattered by atmospheric molecules.

Polarization pattern of sky is invisible to humans, but can be exploited by insects

Page 54: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

54

Path Integration

Random search for food, but almost straight path home when food is found

Page 55: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

55

The desert environment with few landmarks

Page 56: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

56

Lambrinos and his colleagues built a navigation system for Sahabot, based on the ant’s solar compass.Lambrinos, D. et al (1997)

Page 57: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

57

The compound eye of the ant is made up of many ommitidia. It is the upper ommitidia the house the solar compass.

Page 58: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

58

sensors

Polarised light sensors

Page 59: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

59

Page 60: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

60

Sahabot Could use robot to test out different models

of acquiring compass information from the polarized light pattern of the sky

And to test out models of visual landmark navigation

Page 61: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

61

Robot tested in the same environment as ants

But different in size, and method of propulsion

Experiments provide confirmatory evidence for biological hypotheses

Page 62: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

62

Lambrinos et al (1997, 2000) Goal of synthetic methodology, “to develop an

understanding of natural systems by building a robot that mimics some aspects of their sensory and nervous systems and their behaviour”

Biorobotic study can result in innovative engineering designs

And confirm biological hypotheses And generate new biological hypotheses

Page 63: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

63

“This case study illustrates the power of biorobotics

that arises from the close relationship between engineering and biology. On the one hand, insights for innovative engineering designs can be found by analyzing the mechanisms employed by biological agents. On the other hand, biological hypotheses can be confirmed using real world artifacts rather than simulation only, and new biological hypotheses and ideas for new animal experiments can be generated” (Lambrinos et al, 2000).

Page 64: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

64

Task allocation Krieger, M.J.B, and Billeter, J-B (2000) The call of duty:

Self-organised task allocation in a population of up to twelve mobile robots. Robotics and Autonomous Systems, 30, 65-84.

How are tasks allocated in insect colonies? - different castes? (genetic) Or activation thresholds? Relevancy to engineering applications?

Page 65: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

65

Activation thresholds Reaction to stimuli E.g. neglected brood, or corpses of dead

ants – odour increases in strength When it reaches an individual’s threshold

value, they respond E.g. grooming brood, or carrying corpse out

of nest.

Page 66: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

66

Use of activation thresholds for coordination of robot team

- aim to show that implementing different activation thresholds is sufficient for effective task allocation Join waiting line

Exploration of a possible mechanism for self-organisation of social insect societies

Page 67: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

67

Page 68: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

68

Page 69: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

69

Page 70: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

70

Robots’ mission – to collect food items and bring them back to nest to keep nest energy at safe level

Two tasks Remain in nest Forage

Page 71: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

71

Wait in nest Leave nest when nest energy lower than activation

threshold (knowledge of nest energy)

Leave nest No knowledge of nest energy

Look for food items Load food item Return to nest Radio in energy consumption and recharge personal

energy store Unload food item

Page 72: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

72

Experiments Nest energy remained stable – for teams of

3,6 and 9 robots. Decreased for teams of 12 robots.

Experimental manipulations: Group size and food distribution See Figure

Page 73: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

73

Page 74: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

74

Advantages of activation threshold: Decentralised method, can be applied to

variable sized groups Method: equal distribution of activation

thresholds between lower and upper value Effect of other methods? (not explored)

Page 75: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

75

Conclusions of Krieger and Billeter (2000) Individual activation thresholds are an

effective way of allocating tasks Simple mechanism shown to be sufficient

Complex social systems can be regulated in decentralised way

Groups of 3-6 robots did better than single robots Illustrates advantages of transition to social group

Page 76: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

76

Summary What is biorobotics?

The intersection between biology and robotics Robotic model that tests biological hypotheses

Contrast between Biological inspiration Biological modelling

Examples of biorobotic models: Model of cricket phonotaxis (Lund et al 1998) Study of navigation in desert ant (Lambrinos et al 2000) Study of task allocation (Krieger and Billeter, 2000)

Question: What can biorobotic models tell us?

Page 77: Adaptive Robotics COM2110 Autumn Semester 2008 Lecturer: Amanda Sharkey

77

What can biorobotic models tell us? Explicit form of modelling – dealing with real

environmental stimuli Can show that mechanism is sufficient to account for

behaviour But cannot be sure that the model is the mechanism

used in biology (under-determination) Often the challenge is to find simplest mechanism