Collaborative Hunting Method of Multi-AUV in 3-D IoUT: Searching, Tracking, and Encirclement Keeping

In Internet of Underwater Things (IoUT), multiautonomous underwater vehicles (AUVs) can hunt specific targets by performing collaborative hunting tasks. Collaborative hunting task refers to multi-AUV hunts underwater target through collaborative hunting algorithm, which includes target searching, tracking path planning, and encirclement keeping. However, in actual scenarios, this task requires hunter AUVs to search targets independently, and encircle targets continuously, which poses a huge challenge to the collaborative hunting algorithm. To solve this challenge, this article proposes a collaborative hunting algorithm for dynamic ocean targets to ensure collaborative hunting effects. First, this algorithm models the collaborative hunting of hunter AUVs, and designs the kinematic model and detection model of hunter AUVs. According to the 3-D encirclement of hunter AUVs, this algorithm establishes a tracking encirclement metric to evaluate the encirclement keeping of hunter AUVs. In addition, this algorithm adjusts divide areas based on robots initial position (DARP) and bio-inspired neural network (BINN) to achieve collaborative searching in 3-D environment. Then, this algorithm designs an improve crayfish optimization algorithm (ICOA) to obtain tracking paths and hunting actions of hunter AUVs. Specifically, ICOA uses planning space and cubic map to initialize populations, and performs the optimization operation through Levy flight. Extensive simulation results show that the proposed algorithm can complete the collaborative hunting task, and is better than other algorithms in terms of encirclement keeping and tracking path planning.