Posted December 14, 200717 yr My engineering work: "Methods and algorithms for mobile robot positioning" - presentation - gallery - video - website
December 16, 200717 yr Tell me if I have understood this correctly: You have a camera to detect the obstacle, refine the landscape using the calculations on the matrices and plan the route for the robot with an SSSP algorithm that uses the visibility graph? It sounds impressive, but have you planned on trying to make the autonomous agent do some more of this work instead of the computer with web-cam? The distance sensor from the NXT should give you quite good results in such a controlled environment. I can recommend Maja J Mataric's Integration of Representation Into Goal-Driven Behavior-Based Robots: http://www-robotics.usc.edu/~maja/publications/ieeetra92.pdf if you want to explore this field further.
December 18, 200717 yr Author After detection parameters of the objects (robot, obstacle, destination point) i can used the graph of the visibility method to the planning the optimal path. I used Web Camera. Programming language: Visual Basic 6.0 (VB) and Lego Assembler (LASM) (without NQC).
December 18, 200717 yr I'd say good work, even though I'm so dumb when it comes to maths or computers - my interest lies in other areas - I didn't understand any of it, it seems impressive *sweet* In any case *y* *y* *y*
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.