Objectives This paper proposes a multi-source observation-based recovery bucket guidance docking strategy for the reliable recovery observation and motion goal tracking of unmanned surface vessels (USVs).

Methods During the entire docking process, the interference zone of the mother ship's wake is first avoided in order to complete the rough alignment of the USV route; the heading tracking guidance line is then maintained to prepare for terminal docking recovery adjustments; finally, the data obtained by the visual sensor and inertial navigation sensor is filtered and fused to calculate the recovery guidance line, which is then transmitted to the USV. The USV completes the tracking of the terminal guidance line and docking recovery task through its own guidance and control systems. A USV docking recovery system is simultaneously designed on the basis of visual and integrated navigation fusion, the hardware and software of the real boat is independently designed, and lake field tests are conducted to verify the feasibility of the system design and docking strategy.

Results The experimental results show that the success rate of the USV in performing autonomous docking tasks reaches 91.6%. The proposed docking strategy can meet the high-precision docking and recovery requirements of USVs.

Conclusions The findings of this study can provide critical technical support for USV recovery operations.