Full drive auv recovery control system and autonomous recovery method

Abstract

The invention discloses a full-drive AUV recovery control system and an autonomous recovery method, provides a hardware structure for a full-drive AUV recovery docking control system, an autonomous recovery method that uses dynamic positioning to maintain linear tracking and hovering calibration, and A dynamic positioning state estimation method based on adaptive unscented particle filter that can improve the stability of AUV recovery and docking. The adopted adaptive unscented particle filter algorithm uses Saga-Husa to estimate the statistical characteristics of unknown noise, and uses the particle distribution to jointly approximate the state posterior distribution, which complements the advantages of particle filter and UKF algorithm. In the case of external interference, physical characteristics of acceleration and human manipulation, the tracking filtering effect is better, which effectively improves the stability of dynamic positioning and the success rate of autonomous recovery.

Description

technical field [0001] The invention relates to a fully-driven autonomous underwater robot, in particular to a recovery control system and an autonomous recovery method for a fully-driven autonomous underwater robot, belonging to the technical field of robots. Background technique [0002] Autonomous underwater vehicle (AUV) is currently a hot spot in the development of marine engineering technology, and it is playing an increasingly extensive role in marine resource exploration, submarine engineering operations, scientific research and many other aspects. Due to the AUV's own energy limitations, when the AUV completes a certain amount of tasks, it usually needs to be docked and recovered underwater for timely energy supply and data exchange. At the same time, the task for the next period of time will be issued, and the AUV will be deployed underwater again for exploration. and job tasks. [0003] At this stage, AUV recovery generally adopts an underwater recovery device wi...

AI Technical Summary

Problems solved by technology

The particle filter uses the sequential Monte Carlo method to solve the problem that the convergence of the EKF and UKF algorithms drops sharply or even diverges when the nonlinear / non-Gaussian is strong. weight degradation problem

[0005] The application number "201310639638.5" is titled "A recovery system and recovery method for an autonomous underwater vehicle". This method is simple and effective, but it relies on the traction rope shot out by the staff for passive fishing, and its autonomy and concealment are not high.

The application number is "201310141292.6" and the name is "A method for long-distance precise autonomous loading and docking control of an underactuated underwater robot". Due to the limitation of the underactuated structure, this method can only indirectly eliminate lateral errors and depth errors, and cannot hover for positioning and calibration. High requirements on controller and positioning algorithm

Images