Indus river basin’s irrigation system in Pakistan is extremely complex, spanning over 90,000 km. Maintenance and monitoring of this extensive network demand enormous resources. This paper describes the development of a streamlined and low-cost autonomous underwater vehicle (AUV) for the monitoring of irrigation canals including water quality monitoring and water leak detection. The vehicle is a hoveringtype AUV, designed mainly for monitoring irrigation canals, with fully documented design and open source code. It has a length of 0.43 m, and a radius of 0.09 m with a depth rating of 4m. Multiple sensors have been installed onboard the AUV for monitoring water quality parameters including pH, turbidity, total dissolved solids, temperature, and electrical conductivity. A 9-DOF IMU, GY-85, is used, which incorporates an accelerometer, a gyroscope, and a magnetometer. The readings from these three sensors are fused together using direction cosine matrix algorithm, providing the AUV with the roll, pitch and yaw angles. A water pressure sensor, MS5837, gives the depth of the AUV. Four thrusters control the vehicle’s surge and heave, providing 3 DOF. The thrusters are controlled using a proportional-integral-derivative feedback control system, with IMU data and water depth being the controller’s input and the thruster’s speed as the output. The AUV is controlled by a single onboard processor, Arduino Mega 2560.
A flow sensor has been installed beneath the main hull, for monitoring the changes in canal water flow and detecting potential water theft in the irrigation system. For recording underwater sounds, an indigenously developed hydrophone is placed on the AUV. The vehicle also provides information on water quality, providing the capability to identify the potential source(s) of water contamination. Detection of such events can provide useful policy inputs for improving irrigation efficiency and reducing water contamination. The AUV, being low cost and of small size, is suitable for autonomous maneuvering and water quality monitoring in the irrigation canals and can be used for monitoring irrigation networks at a large scale.