File Exchange

image thumbnail

Virtual Robot Demo

version 1.0 (20.7 KB) by

Test navigation algorithms in a virtual robot simulator



View License

This program demonstrates obstacle avoidance/navigation a simulated two-wheel robotics platform. I simulate the drive train with the differential steering model, which is simple, but very useful model for robotic actuation. This type of drive system can drive straight, trace an arc, and turn in place.
The simulated robot makes its way around a simulated environment represented by a .png file where black represents walls, and white represents flat, clear terrain. The “environment” a black and white image. For testing, I include an organic-looking obstacle course and an “office” with square walls and corners.

The simulated robot wanders around the environment and covers most of the area without hitting walls. This is accomplished with this algorithm:

1. Go straight unless sensors are blocked
2. If one sensor is blocked, turn away from it
3. If both are blocked, turn towards the farthest
4. Turn until both sensors are free
5. Continue going straight (with random arcs)

Check this site for videos of the demo and more robotics, and computer vision projects.

[ ]

Comments and Ratings (5)

nasma salam

I wip exicution this programe but the cours.png it do disply not whay?

I have tried changing the code to add two robots instead of one. Struggling ... Any hints?


shlomix (view profile)

I like this code and I'm using the simulation within it as a to make an evolutionary nural network controler for a Khepera robot. For now on I found one bug:
In rangefinder function [dimy, dimx] = size(course); should be [dimx, dimy] = size(course); . It's not important for symmetrical PNG's but for nonsimetrical the robot is able to go through walls - nice but not desirable.

jggg kjhh


jia qing




Added BSD License

MATLAB Release
MATLAB 7.4 (R2007a)

Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.

» Watch video

Win prizes and improve your MATLAB skills

Play today