16-311-Q I NTRODUCTION TO R OBOTICS L ECTURE 13: R EACTIVE CONTROLLERS B EHAVIOR -B ASED C ONTROL N AVIGATION & O BSTACLE AVOIDANCE I NSTRUCTOR : G IANNI A. D I C ARO
A CONTROLLER FOR PHOTOTAXIS BEHAVIORS? • The robot has two sensors, to the left and the right side respectively • Each sensor reports the amount of detected light • If the amount of light, averaged from the two sensors, is less than L, than the robot exhibits phototaxis: it heads toward the light • If the amount of light, averaged from the two sensors, is greater than L, than the robot exhibits anti-phototaxis: it escapes from the light We need to switch between two opposite behaviors What type of feedback controller should we use? 2
BANG-BANG CONTROLLERS: BINARY FEEDBACK Output(1, Gain) Output(2, Gain) • Two-states systems >=Threshold • State transitions happen < Threshold depending on a threshold value • The action at each state depends State 1 State 2 on a fixed gain parameter G • Hysteresis can be helpful (in >= Threshold general) to avoid over reacting < Threshold Wall following State 1: Too near - State 2: Too far Threshold: Binary output from a proximity sensor (wall y/n) Gain parameter: The ICR radius for moving to/away the wall Hysteresis: In absence of a distance measure, a time interval 3
HYSTERESIS Output(1, Gain) Output(2, Gain) Output >=Threshold Out(1, Gain) < Threshold Input ≥ Θ ≦ Θ Δ State 1 State 2 Θ Θ Input Θ Δ ≥ Θ >= Threshold Input < Θ < Threshold Out(2, Gain) Output No hysteresis Out(1, Gain) Bang–bang controls with ≥ Θ Input ≦ Θ - Δ hysteresis provide optimal controls With hysteresis in some cases, although they are Θ Θ Input often implemented just because of Θ - Δ their simplicity or when binary Input ≥ Θ behaviors are required Θ Out(2, Gain) 4
B R A I T E N B E R G V E H I C L E S Basic vehicle ∼ One single neuron / motor that Braitenberg vehicle basics fires proportionally to the Implementation as simple electronic circuits external input (from the sensor ) • Sensors directly connect to motors Activation can be + or - • Can increase or decrease motor activity Two or more “neurons” can be • Behavior depends on put together … • geometrical arrangement of sensors • geometrical arrangement of motors • cross connection of sensors • Emergent behaviors • Competing stimulation / inhibition creates complex actions 5
B A S I C T Y P E S Basic three types of Braitenberg vehicles • Sensors stimulate same-side motors • Sensors stimulate opposite motor • Sensors go to both (all) motors, some stimulating, some inhibiting 6
O B S TA C L E AV O I D A N C E • Sensors generate proximity value • The closer to an object, the higher the sensor output • Obstacle left • Obstacle right • Obstacle center 7
L I G H T S E E K I N G • Sensors generate light value • The more light, the higher the sensor output • Obstacle left • Obstacle right • Obstacle center 8
C O M P L E X B E H AV I O R S Braitenberg: Light seeking + obstacle avoidance Depending on weights, di ff erent behaviors • Following a path • Docking to a charger station • Coordinated moments of a swarm of robots • ... and many more 9
F R O M A N O B S E R V E R P O I N T O F V I E W … Out of simple rules (and possibly imprecisions in actuations that naturally break symmetries) a number of di ff erent behaviors can be displayed that, to an external observer, could look like expressions of some personality, emotion, social interaction … 10
F O R M O R E I N F O … A nice video talking in general about Braitenberg’s book and his vehicles + nice small robots programmed to act as the vehicles and show a complex world … https://www.youtube.com/watch?v=A-fxij3zM7g 11
REACTIVE PARADIGM FOR ROBOT CONTROL: DON’T THINK, REACT! Starting from the mid 1980s, a number of di ff erent views (mostly bio-inspired ) and approaches were developed and employed in robotics (and in AI, that started moving from symbolic to sub-symbolic / neural models) Deliberative Top-down paradigm problem solving Functional Requires a Look Internal pipeline closed world ahead models Concurrent React Model Works in an modules to inputs free open world Bottom-up Reactive problem solving paradigm 12
ARCHITECTURE BASED ON REACTIVE PARADIGM Sense-Act Transfer rules Ethological view ( Behavior ): Behaviors Direct mapping of sensory inputs to a pattern of motor actions that are then Rule 1 used to achieve a task Rule 2 Mathematical view ( Function ): A transfer function, transforming Rule 3 sensory inputs into actuator commands Rule n Sense Act Concurrent mode Environment vs. Sequential mode Vertical decomposition vs. Act Sense Model Plan Horizontal decomposition Environment 13
EXAMPLES OF SENSE-ACT RULES Identify Objects Monitor Changes Explore Wander Avoid Objects Build Map Find Path Environment Sense Act Track Person Follow Wall Avoid objects Environment Sense Act 14
QUESTIONS TO CLARIFY 1. Where the Reactive paradigm finds its roots? 2. What is the exact nature/characteristic of the SENSE-ACT rules? 3. How the ACTion output from the different rules is arbitrated to result into a single, coherent Action command to the Effectors? Good reference books, plenty of ideas and discussions on the Reactive / Behavioral paradigm: R. Arkin, Behavior-Based Robotics , MIT Press, 1998 R. Murphy, An Introduction to AI Robotics , MIT Press, 2000 M. Mataric, The Robotics Primer, MIT Press, 2007 J. Jones, A Practical Guide to Behavior-based Robotics , McGraw-Hill, 2004 15
THE BIOLOGICAL ROOTS 1. Where the Reactive paradigm finds its roots? Dissatisfaction with the limitations of the Deliberative approach Many “simple” animals exhibit individual and Animals live in an open world , collective intelligent behavior yet have virtually no and roboticists would like to overcome the brain. Therefore, they must be doing something closed world assumption to manage world’s representation complexity! Historical track: Cybernetics (back to the 40’s, N. Wiener, G. Walter’s Turtoises ), V. Braintenberg’s (conceptual) vehicles based on direct SENSE-ACT 16
REACTIVE RULES AS BEHAVIORS 2. What is the exact nature/characteristic of the SENSE-ACT rules? A fundamental building block of Releaser natural intelligence is a behavior : a mapping of sensory inputs to a Sensor Pattern of pattern of motor actions, Behavior Inputs Motor Action which then are used to achieve a task Ethology: study of Sensor input : Water source detected animal behaviors Releaser: Giraffe is thirsty Releaser: No predators Task: Drinking Action pattern : Move head checking for predators Put legs in right position, Lower the neck Adjust legs position Drink rapidly Neck up and check surroundings 17
CATEGORIZATION OF (ANIMAL) BEHAVIORS Reflexive behaviors: Stimulus-response (S-R), “hardwired” behaviors. The stimulus is directly connected to the motor action to produce the fastest response time. No cognition: if you sense it, you do it! Arctic tern feeding: When hungry, babies peck at the closest red blob Reactive behaviors: Increase of complexity Learned, and then consolidated to where they can be executed without conscious thought, but can be changed by conscious thought. Hunting “Muscle memory” Conscious behaviors: Deliberative , requiring conscious thought, possibly Assembly a robot? combining previously developed behaviors In ethology , a reactive In robotics , the word reactive behavior means a learned (mostly) connotes a reflexive Source: R. Murphy, AI Robotics , MIT Press, 2000 behavior or a skill behavior. R. Arkin, Behavior-Based Robotics , MIT Press, 1998 18
CATEGORIES OF REFLEXIVE (ANIMAL) BEHAVIORS Fixed-action Reflexes Taxes patterns Response: to move to a Stimulus t 1 ——— t 2 Stimulus t 1 — t 2 particular orientation Response t 1 ——— t 2 Response t 1 ———— t 3 t 2 marked by a stimulus intensity Short/Instantaneous Persistent 100 Value Axis Response Phototaxis 50 0 Keep fleeing after Stimulus intensity AprilMayJune July predator detection Category Axis Chemotaxis 19
FROM ANIMAL TO ROBOT BEHAVIORS Why all this discussion about animal behaviors when we are (mainly) interested to robot behaviors ? A. Ethology / Animal behaviors have been, are, and will be a major source of inspiration to design effective robotic behaviors! B. As a matter of fact, in terms of taxonomy, design principles, architectural organization, the reactive paradigm in robotics has mirrored very much the results coming from the ethology field C. What we have discussed so far has been translated /adapted into a number of different views of the reactive paradigm: Subsumption, Motor schema, Potential fields, Behavior-based control, … 20
A PRACTICAL EXAMPLE: OBJECT COLLECTION Storage Task specification: A • The robot searches for type A objects A A • When an A object is found, it has to be brought (by pushing) at a storing location identified by a bright light A A • The robot has to collect as many A objects as possible A • Other objects cannot be pushed • The environment can feature walls A Robot The robot: A • Frontal IR emitters / detectors • Light sensors • Frontal bumper Does it look like as a simple task • Two standard wheels to program/control? 21
Recommend
More recommend
