Optimization of swarm robotic inspection - Micro with different learning schemes Course project presentation Swarm Intelligence EPFL Lausanne Switzerland February the 6 2006 Vincent Cattin & Nils Raning
The blade experiment A swarm of robots inspect a regular environment to detect problems on blade.
Our microscopic model Our finite state machine: p virgin (k) T blade Circle virgin blade T wait Search Serve as Beacon (1/T half )p beacon (k) Circle T blade inspected blade p inspected (k) p wall p robot (N0-1) T wall T robot Avoid wall Avoid robot
Our algorithms: -PSO -GA -In-line search
PSO Particle swarm optimization -Particles start at random position with a random speed -They are attracted by the best place they have been and the best place found by the swarm.
Genetic Algorithm
In-line search
Results GA for 5 robots 1000 900 -T-start > nbr of steps 800 -T-start not usefull with 700 few robots Tmax Tstart -Good optimization with 600 Twait all algorithm 500 Steps 400 300 200 100 0
Discution R = # robots, B = # blades. 1 beacon = -1 robot to explore and +1 blade « beaconned » 1 beaconned blade = avg 50% less time spend on the blade => 50%/B less exploration needed 100%/R less explorer => to win we need R > 2B => we need LOTS of robots to need beacon might be usefull in not uniform spatial distribution of robots not supported by our model
Recommend
More recommend
