An Overview of Robot Behavior Control

41
Jaroslaw Kutylowski 1 HEINZ NIXDORF INSTITUT Universität Paderborn Algorithmen und Komplexität An Overview of Robot Behavior Control with insight into AI-based and algorithm-based approaches

description

An Overview of Robot Behavior Control. with insight into AI-based and algorithm-based approaches. Agenda. What is this talk going to cover? What is behavior? Behavior control Basic control strategies Advantages and disadvantages of these strategies Hybrid strategies Behavior-based control - PowerPoint PPT Presentation

Transcript of An Overview of Robot Behavior Control

Page 1: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 1

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

An Overview of Robot Behavior Control

with insight into AI-based and algorithm-based approaches

Page 2: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 2

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und KomplexitätAgenda

What is this talk going to cover?

• What is behavior?

• Behavior control- Basic control strategies- Advantages and disadvantages of these strategies- Hybrid strategies

• Behavior-based control

• Deliberation-based control

• Hybrid strategies

• Final remarks

Page 3: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 3

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und KomplexitätWhat is behavior?

• every robot has a goal• how to accomplish this goal?• good readings from sensors and good control of movement do not

suffice

we need proper decision-making

Page 4: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 4

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior control

High-level behavioral algorithms

Low-level basic algorithms

Physics

What is behavior?

• movement control

• sensor control

• what to do on sensor input• how to coordinate with teammates

• navigation• exploration• etc.

Page 5: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 5

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

What is behavior?Behavior control

Input:• sensory data• history of behavior• information from teammates• information about opponent

Output:• what to do next?

– where to go– where to look– what to send to teammates

Behavior control

High-level behavioral algorithms

Low-level basic algorithms

Physics

We will look at different methods for decision making and following

these decisions

Page 6: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 6

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

What is behavior?High-level behavioral algorithms

Most prominent problems• navigation to a point, with obstacles • exploring unknown terrain• task allocation

Behavior control

High-level behavioral algorithms

Low-level basic algorithms

Physics

Many research done in this area

We will review some of the results

Page 7: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 7

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

What is behavior?Low-level basic algorithms

Behavior control

High-level behavioral algorithms

Low-level basic algorithms

Physics

Typical problems• how to walk• how to read sensor input• how to evaluate visual sensory input

These are problems which we won’t discuss

Page 8: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 8

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior controlbasic strategies

Two main approaches to behavior control:• Behavior-based control (reactive)

– “world is the world’s best model”– simple actions as reactions to environment– complex behaviors emerge from simple ones– stateless– no communication between teammates, only observation– inspired biologically– emerging from the AI community

• Deliberation-based control– careful planning of actions– maintaining state and synchronizing it with the environment– complex behaviors planned in advance– communication with teammates– prediction of opponent’s behavior– emerging from the algorithmic community

Page 9: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 9

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior controldiscussion on behavior-based control

Advantages:• Simple controller, suitable for architectures with low performance• Easy implementation leading to rapid development• Easy to test and debug• Should adapt well to changing environmental conditions• Fast reaction time, well suited for dynamically changing situations e.g.

(e.g. robot-soccer)• Provable low-level properties (collision-avoidance etc.)

Disadvantages:• Emergent behavior is impossible to predict• No provable properties about emergent behavior• Not suitable very well to less dynamic situations where goals are

achieved in a long term (e.g. UGV navigation)

Page 10: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 10

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior controldiscussion on deliberation-based control

Advantages:• Possibility to plan in advance for long term behavior• Complex behaviors are precisely defined and provable• Can take advantage of communication with mates• Possibility of learning and thus predicting the moves of the opponent

Disadvantages:• High hardware requirements (computationally intensive algorithms)• Possibility of loss of synchronization between internal state and

environment• Problems hard to solve and implement• Can react too slow in very dynamic situations (e.g. robot-soccer)

Page 11: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 11

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior controlhybrid strategies

The two basic strategies can be combined to hybrid ones• Basic behavior controlled by behavior-based strategies (low-level)• Deliberation-based methods define a high-level strategy

Advantages of both strategies can be combined

Page 12: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 12

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlagenda

• Typical methods:– Simple state machines and how to define them

– Potential fields method

– Formation control

Page 13: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 13

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlsimple state-machines

Most popular method of behavior control in dynamical systems

Used by GermanTeam 2002 and later

Sample definition: Goalie

Goalie-before-kickoff

Goalie-playing

Return-togoal

Position-inside-goal

stand go-to-point

kick

go-to-ball

Page 14: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 14

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlXABSL for defining behavior rules

Instead of defining behavioral aspects of software in plain code, usage of meta-languages

Software engineering defines UML, Petri-nets, high-level scripting etc. for modeling of behavior

XABSL (extensible agent behavior specification language) is defined by the German team

• Syntax based on XML• Defines a state-automaton• Language constructs typical for a structural language (if, conditions)• Constructs for easy operation on the state-automaton (transitions)

Basic behaviors like “go-to-ball” defined in low-level language

Page 15: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 15

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based control XABSL for defining behavior rules

XABSL is transformed into Intermediate Code, which is executed on the AIBO by a low-level virtual machine

AIBO behaves according to the definitions given in XABSL, acting as a state-automaton

Page 16: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 16

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based control decisions in state machines

Sometimes decisions between certain behavior options must be made

These are based on evaluating utility functions for possible options

These utility functions can be influenced by non-determinism

Page 17: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 17

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlPotential fields method

Objects either attract or repulse the robot

These forces constitute the potential field

Forces in the field are summed according to physical rules, so that one obtains the resultant force

The resultant force indicates the movement direction of the robot, optionally the force strength determines the movement speed

Advantages:• Smooth movement• Elegant solution, very easy to describe

Page 18: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 18

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlPotential fields method

Ball

Robot

Opponent

Repulsive force

induced by opponent

robot

Attractive induced by

ball

Calculated resultant force,

direction of movement

Page 19: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 19

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlPotential fields method – problems

Local minima

Ball

Robot

Ball

Robot

Page 20: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 20

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlPotential fields method – problems

No passage between close objects

Ball Robot

Page 21: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 21

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlPotential fields method – problems

Oscillations

Ball

Robot

Ball

Robot

Page 22: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 22

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlformation control

Formation control is important for terrain traversal, soccer …

Four robots travel in a predefined formation

Column Line Diamond

Robots compute position and positions of others

Own formation position is calculated basing on• leader position• neighbor position• unit-center position

Page 23: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 23

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Behavior-based controlformation control

Robot tries to maintain formation, by staying inside of the dead zone

Inside of dead zone no additional formation maintaining performed

Inside of controlled zone speed vector into dead zone linearly dependent on distance from dead zone

If obstacles occur, the avoidance gains priority As soon obstacle is surrounded, the robot tries to get into formation Can be realized using potential field, with dead zone attracting and

obstacles repelling

Dead zone

Controlled zone

Page 24: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 24

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based controlagenda

• Typical methods:– Case based reasoning

– Hidden Markov Models

• Algorithmic approaches– task allocation

– navigation to a point, with obstacles

Page 25: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 25

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based controlcase based reasoning

During soccer play similar situations can occur quite often

Case based reasoning allows a player to store the behavior of opponents and use it when a similar situation occurs once again

Sample:

Goal

Robot with ball

Opponent

Page 26: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 26

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based controlcase based reasoning

Advantages:• Opponent behavior can be analyzed and player can adapt to its

strategies

Disadvantages:• If opponent uses similar techniques, than the two CBR instances fight

against each other, returning improper forecasts• No provable results• Highly memory and computational intensive• Learning process is needed

May be advantageous against simple opponents, but has no provable properties and fails against “intelligent” opponents

Page 27: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 27

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based controlHidden Markov Model method

As in CBR, the goal is to predict the behavior of the opponent

The HMM method:• assume that the opponent has a state machine and uses a set of

common behaviors, like go-to-ball, intercept-ball …• for each behavior we define a model, which is a state machine with

probabilities for transition from state to state• for every possible observation the model contains a probability that it

occurs in a certain state• we cannot directly observe the state of the opponent so we

instantiate HMM behavior models and look whether their execution matches the observations

• thus we obtain probabilities that the opponent is in a certain state of a certain behavior

Page 28: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 28

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based controlHidden Markov Model method

Observations are

• Distance of robot to ball

• Robot ball manipulation

• Distance of robot to goal …

The most interesting question is about the value of

Knowing the probability, we can derive some information about future behavior of opponent

],...,,|Pr[ 21 tit oooOsS

Page 29: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 29

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodstask allocation

Task allocation is important when coordination of robots is needed

With robot soccer task allocation is mostly reduced to role assignment (first forward, supporting forward, defender)

Lot of research on multiprocessor task scheduling and similar assignments, which can be often translated to multi-robot scenarios

Models utilized for robot task scheduling:• Robots are heterogeneous • Tasks require specific skills• Tasks appear online• Communication is expensive and thus must be minimized• Computation power is sparse

Page 30: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 30

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodstask allocation

We look for efficient, online and distributed approximations for task-allocation

Taxonomy of task allocation problems:• ST-SR – single-task robots, single-robot tasks• ST-MR – single-task robots, multi-robot tasks• MT-SR – multi-task robots, single-robot tasks• MT-MR – multi-task robots, multi-robot tasks

uncommon

Page 31: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 31

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodstask allocation – ST-SR setting

Model

• Set M of workers, s. t. |M| = m

• Set N of jobs, s. t. |N| = n jobs, with a weight wj for each job

• skill rating, which defines the fitness of a worker for a job:

We want to find such an assignment of workers to jobs, s. t. a sum of the combination of utility function and job weight is maximized

Centralized ILP solvable by Hungarian Method gives runtime of O(mn2), but needs about n2 messages to be exchanged

Distributed auction mechanisms achieve the same task with only O(n) messages

RMNU :

Page 32: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 32

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodstask allocation – online ST-SR setting

The previous model assumed an offline-setting

In reality the online version is much more likely to occur

BLE algorithm:• If any robot is unassigned, find the robot-task pair with highest utility

and weight• Assign this robot to this task• Go on

This greedy strategy is 2-competitive to the optimal offline algorithm

Page 33: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 33

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodstask allocation – ST-MR setting

Also known as coalition formation

Now each job might require a specific skill which is possessed only by some robots

Transforming the coalition formation problem to SPP:• Let E be a set of all tasks and robots• Let F be a family of all robot-task pairs• u(f), where f is a set from F, is the utility for robot-task pair

SPP • Finite set E• Family F of subsets of E• Utility function u: F→R+

Find a maximum-utility family X of elements in F, s.t. X is a partition of E

Page 34: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 34

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodstask allocation – ST-MR setting

SPP is NP-complete

But there are heuristics and approximations which give good practical results

Unfortunately these methods do not have a guaranteed approximation ratio, they only report how far the constructed solution is from the optimum for a particular problem instance

Page 35: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 35

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodsnavigation to a point

Model:• The robot should get from a source position to a target position

traveling the smallest possible distance• There are obstacles with unknown position and size

Different assumptions about the abilities of sensors may be made• Visual sensors• Touch sensors

Important measures:• Ratio of distance obtained by algorithm and the optimum• Distance taking into account the sizes of obstacles

Page 36: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 36

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodsnavigation – D* algorithm

Model• Finite undirected graph G(V,E), most often a grid• Edge blocking • The edge blocking is unknown to the algorithm• The blocked edges cannot be traverse• Blocked edges can be detected only at adjacent vertices

D* algorithm• Assume that all the unknown terrain contains no blocked edges• Find shortest path• Try to go on this path• On blocked edges update terrain map, calculate new path

EB

Page 37: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 37

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodsnavigation – D* algorithm

Sample edge-blocked graph

S

E

Page 38: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 38

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Deliberation-based methodsnavigation – D* algorithm

Performance of D*• Lower bound on competitive ratio

• Upper bound on competitive ratio

Lower bound construction

Page 39: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 39

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und KomplexitätHybrid strategies

Two layers of execution

• The lower runs with reactive behavior-based methods

• The upper runs with deliberative methods

The lower layer assures fast reactions, obstacle avoidance etc. and can basically function without the help of the upper layer

The upper layer provides additional support to the lower layer, by analyzing the situation (e.g. case based reasoning) and giving “hints” to the lower layer

The hints are only supportive for the working of the behavior-based methods, i.e. they can be (partially) ignored

The hint can be modeled as a slight influence on a utility function of executing an option

Page 40: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 40

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und KomplexitätFinal remarks

What you should remember

• Two basic strategies for behavior control

• No clear indication which one is best

• Many research in both areas, with deliberative having more strict proofs and behavior-based having more practical realization

Today’s results aren’t great Practical realizations more often use simpler methods – there is a gap

between the theoretical results and their implementation

Page 41: An Overview of Robot Behavior Control

Jaroslaw Kutylowski 41

HEINZ NIXDORF INSTITUTUniversität Paderborn

Algorithmen und Komplexität

Jaroslaw KutylowskiHeinz Nixdorf Institut& Institut für InformatikUniversität PaderbornFürstenallee 1133102 Paderborn

Tel.: 0 52 51/60 64 68Fax: 0 52 51/62 64 82E-Mail: [email protected]://www.upb.de/cs/ag-madh

Thank you for your attention! Thank you for your attention!