25-01-2011, 11:32 AM
[attachment=8425]
Aarathi Vikas
S7 CSE
INTRODUCTION
Self healing robots have ability to adapt to minor injuries
A robot is able to infer its own morphology through self directed exploration
The concept used is multiple competing internal models and generation of actions
THE STARFISH ROBOT-CHARACTERIZING THE TARGET SYSTEM
The target system is a quadrapedal, articulated robot
It consist of a rectangular body and four legs attached to it with a hinge joint
Each leg has a upper and lower leg attached together with a hinge joint
All eight hinge joints are actuated with high torque servomotors
Servo drives capable of producing max 200 ounceinches of torque and 60o per second speed
Robot can automatically synthesize predictive models of its own topology
Equipped with suite of different sensors
SELF MODELLING BRIEFLY
Robots depend on internal maps and sensory data to update their location
In a changing environment the robot has to navigate with uncertainity
Situation becomes even worse if the robot’s own shape and configuration can change
PHASES IN SELF HEALING
Self –Model synthesis
Exploratory Action synthesis
Target Behaviour synthesis
ALGORITHM
Estimation-exploration algorithm is used
It has 2 functions: damage hypothesis evolution and controller evolution
It maintains a database which stores pairs of data
STEPS INVOLVED….
Exploration phase: Controller evolution
Physical robot failure
Estimation Phase: Damage hypothesis evolution
EXPERIMENTAL SETUP
The quadrapedal robot has eight degrees of freedom
Two one degree-of-freedom rotational joints per leg:one at the shoulder and one at knee
It has 4 binary touch sensors and also 4 angle sensors
All joints are actuated by a torsional motor
THE CONTROLLERS
Robots are controlled by a neural network
There are 3 layers in the neural network-input layer,hidden layer,output layer