Autonomous multi-sensor platform with stereo vision

This paper is describing a different way to investigate the terrain topography: stereo images. This is the closest way of terrain investigation to the human sight. By gathering as much information from images the need for other sensors on the robot will be no more. This is an obvious fact since for human more than 90% of information about the surrounding environment is coming from sight. An equipment to capture and process stereo images had been build and the software was created to prove the accuracy of this method to detect and avoid obstacles and to investigate the terrain topography for uncharted areas. To extract the information needed for the navigation from the stereo optic images, complex algorithms of shape recognisance, pattern mach and cross correlation will be involved together with image calibration and sharpening. Because the processing power requirements will be huge methods to decrease the computation need will be used to increase the processing speed The sensor and control board are build on one PCB. High power digital signal had to been as far as possible from the high accuracy analog inputs. The magnetic compass proves a very high accuracy: 0.5 degrees. For fast movements, the gyroscopes can give better measurement,. The triaxial accelerometer is a poor way to measure movement. It is mainly used as an inclinometer to help the algorithm of the electronic compass. For the measurement of speed and distance, an odometer is used. It is having a very good resolution, few millimeters, but it is insensitive to car sliding and mechanical interconnection imperfection. For these the accelerometer can be used to improve the result. The GPS results are used for less accurate measurements since its accuracy of positioning is around 5-10 m. The information from all sensors is centralized into a powerful MCU. The MCU alone can be used for controlling the robot movement on a simple route. For more complex situations, this acquisition module is connected to a small- PC able to conduct more complex data analysis.