Landmines cause loss of lives and slow down the economical development in Egypt due to the vast amount of unused land that could either be used in agriculture or in tourism.

For this reason we chose our B.Sc. (Elec. Eng, Computer & Systems Dept.) project to be a mine detecting mobile robot using state of the art technology. The robot uses an advanced java based micro-controller that acts as the brain for the robot. The robot uses a variety of sensors to sense its orientation, location and surroundings. These sensors give readings, which are used by the micro-controller to give suitable output for motors using fuzzy control. The robot is supported by PC software for intensive processing and large memory requirements of AI computations. This is achieved through wireless communication using a proprietary data-communication protocol between robot and PC software.

The purpose of the AI computations is to first formulate a safe map for all possible paths the robot can use. Then the following operation is calculating the optimum cost path using the A* search algorithm between source and destination.

PC software is also used for robot remote control and for the displaying of updated robot and terrain status. The robot status is represented by robot location, orientation and battery status. The terrain status is represented by mine and obstacle locations.

Image processing of GPR images and neural networks have also been applied in identifying buried object type in GPR images. The result is the fusion of two sensors in mine detection (GPR and metal detector).

Through the use of such advanced automated mobile robots we can accomplish two things. On one hand, human losses in mine detection can be eliminated and on the other hand we can gain economically from making use of the large amounts of mine-wasted land in Egypt.