Laser Guided Vehicle Java and MATLAB for Control
Bengt Gunnarsson · KTH Publication Database DiVA (KTH Royal Institute of Technology) · 2005
This study addresses the generic problem of navigating a robot between obstacles. The experimental case tested here is driving through doorways, which can be seen as a special case of a junction between corridors. The robot used was the MICA wheelchair. On-board there is a scanning laser tilted downwards, a fiber optic gyro and odometric encoders. From the laser scan walls and the door are found using the Hough/Radon transform. The results with the control law used gave a standard deviation of 1.4 cm for the lateral error and 2.2 degrees in angle. This can be improved by tuning of control law. The programming was mainly done in Java. Java methods can be integrated in a MATLAB script. MATLAB and Java together was found to be an excellent test-bed for designing and testing algorithms. All computations was done in real-time. A long term goal in robotics is to fuse together many sensors into one platform and make a multipurpose vehicle.