| Popis: |
Unmanned surface vehicle(USV), as an intelligent surface tool, can realize basic functions such as navigation control, as well as surface target recognition and perception. However, with the diversification of engineering requirements, such as search and rescue missions oriented to wrecked freighter, unknown ship feature recognition, and wrecked aircraft, these basic functions fail to complete the task. Therefore, it is necessary to develop a method to track the target stably and complete 360-degree scanning. In this paper, based on the perceptual target information, a target surround tracking method was proposed for target detection. After estimating the target motion state from the sensing information, the virtual target point tracking algorithm was used to calculate the desired heading and velocity. The heading controller and speed controller were used to control the position and attitude of the USV. At the same time, a re-planning and tracking strategy was added to adapt to the complex environment of the water surface and the error disturbance of the information provided by the sensing equipment. The USV model identification method was used to obtain the model and build a simulation platform for simulation verification. The actual flight test proves the rationality of the algorithm design. |
