Doctor of Philosophy
School of Electrical, Computer and Telecommunications Engineering
Shahri, Alireza Mohammad, Intelligent impact control in anti-personnel mine detection, Doctor of Philosophy thesis, School of Electrical, Computer and Telecommunications Engineering, University of Wollongong, 1998. http://ro.uow.edu.au/theses/1952
According to the International Committee of the Red Cross and Red Crescent Societies (ICRC), there are approximately 110 million land-mines scattered around the world in 64 countries. There are also as many mines in the stockpiles around the world waiting to be deployed. As the result of explosions of mines, around 2000 people are killed or maimed monthly. The victims are mostly civilians including women and children who are trapped by mines after the end of hostilities.
At present mine clearance mostly takes place manually. Unfortunately, on average for every 5000 mines cleared one mine clearer is killed. Using the current approach, it would take more than 1,100 years to clear the mines planted in the world at a cost of US$33 billion.
The main focus of this thesis is to investigate the feasibility of developing a robotic arm to detect buried Anti-Personnel (AP) mines in the field and hence to overcome the above mentioned problems. The hand-prodding manual demining procedure using a bayonet is simulated using a single degree force sensing robotic arm.
The robotic device inserts a bayonet into the soil. A strategy to control the bayonet is developed by modelling the dynamics of the manipulator and environment, while adapting for variation in the stiffness sensed by the bayonet when it comes in contact with the mine or any other object in the soil. An explicit impact control scheme is applied as the main control scheme, while three different intelligent control methods are designed to deal with uncertainties and varying parameters of the environment.
An analytical object recognition algorithm based on multiple prodding is developed. A multi-probe mine detection mechanism with the ability to detect a mine faster than a single-probe mechanism is also proposed.
All the developed algorithms are validated through computer simulation and experimented work. The intelligent control algorithms have outperformed the conventional controllers in all of the case studies. They have also produced performances acceptable for demining operation.