Mechanical design: The robot must handle the special limitations of the space confinement and network structure of underground cables.
The control strategy: It includes failure tracking, collision avoidance, and path planning. The control system receives initial commands from the operator for the global tasks, and small tasks are carried out automatically.
The signal-processing algorithm: The robot requires considerable computational resources to be adaptive and flexible. This is highly problematic because of the limited size of the robot, especially for underground applications. It also involves allocation between local and remote signal processing.
Communication: The communication modules exchange data between the host computer and the mobile robot. This includes data originating from different streams on both sides of communication link.
The prototype of the autonomous mobile robotic platform is designed to satify the following requirements:
Mechanical design: The robotic platform consists of two modular segments coupled by a freely rotating joint. Additional segments may be added when functionality evolves. The platform is height constrained by the tunnel environment, and consequently has a very low profile design. The front and rear modules contain hourglass shaped wheels with 12V DC motors to propel the platform along the cable.
Distributed control: The control board consists of three Atmel micro-controllers. Each has specific functions related to the robot's operation, which greatly increases the performance of the robot. The control system coordinates the collection of data, communications, movements and autonomous functionality.
Distributed signal-processing: In order to solve the problem of allocation between local and remote signal processing, the digital signal processor (DSP) board is introduced into the local robot system. The high performance of the DSP board collaborating with a 14 bits, 8MSPS analog-to-digital converter makes the local preliminary signal processing possible. An algorithm to sample and analyze data from the acoustic sensor is implmented to detect partial discharges.