Distributed Adaptive Neural Network Control Applied to a Formation Tracking of a Group of Low-Cost Underwater Drones in Hazardous Environments

Abstract

This paper addresses a formation tracking problem of multiple low-cost underwater drones by implementing distributed adaptive neural network control (DANNC). It is based on a leader-follower architecture to operate in hazardous environments. First, unknown parameters of underwater vehicle dynamics, which are important requirements for real-world applications, are approximated by a neural network using a radial basis function. More specifically, those parameters are only calculated by local information, which can be obtained by an on-board camera without using an external positioning system. Secondly, a potential function is employed to ensure there is no collision between the underwater drones. We then propose a desired configuration of a group of unmanned underwater vehicles (UUVs) as a time-variant function so that they can quickly change their shape between them to facilitate the crossing in a narrow area. Finally, three UUVs, based on a robot operating system (ROS) platform, are used to emphasize the realistic low-cost aspect of underwater drones. The proposed approach is validated by evaluating in different experimental scenarios.

Seabed Surveillance and Underwater Structures Inspection with Remotely Operated Vehicle − Power Ray

“…We recognize that in the future it will be necessary to improve the underwater georeferencing of data collection to estimate the size of the organisms and underwater structures' location. Few of the researchers have dealt with underwater drone positioning and navigation (Capocci et al, 2017;Pham et al, 2020). Moreover, the success of using unmanned aquatic drones to inspect underwater infrastructure or to study underwater ecosystems has been observed to be highly dependent on the turbidity/ transparency of the water.…”

Experimental Investigation of Relative Localization Estimation in a Coordinated Formation Control of Low-Cost Underwater Drones

“…A distributed adaptive training tracking control was then designed based on the consensus tracking protocol and the feedback linearization technique, which is given by: where is the return control gain and is the updated weight. Readers can find the details about formation tracking control in our previous research [ 14 , 32 ].…”

“…Using EKF theory and the vision-based estimation presented above, combined with our previous theoretical studies [ 14 ], we introduce the implementation of relative location estimation for real robots in the next section.…”

“…In previous work, we introduced a control architecture [ 14 ]. It consists of four components, which are the formation tracking control input, , the robust control input, , the neural network control input, , and the collision avoidance, (see Figure 4 ).…”

“…Interested readers are referred to survey papers [ 12 , 13 ] for excellent reviews of the progress made before 2017 in the multi-agent coordination problem. In our previous research [ 14 ], we also demonstrated and simulated on ROS/Gazebo a complete control architecture for a group of AUVs that can maintain formation, as well as avoid collisions and change the distance among them when crossing narrow areas.…”