Search

Path planning algorithm is one of the core algorithms for autonomous underwater vehicle (AUV) to complete autonomous underwater cruise. In this paper, two kinds of AUV path planning algorithms based on environment modeling and path searching are reviewed. The domestic and foreign research status of environmental modeling methods such as grid method, visibility graph method and Voronoi diagram method is expounded, and their advantages and disadvantages are pointed out. In addition, the research results of predecessors in artificial potential field algorithm, fast marching algorithm, A* algorithm and other traditional path search algorithms are analyzed, and compared with the recent intelligent bionics path search algorithms such as particle swarm optimization algorithm, ant colony algorithm, genetic algorithm and artificial neural network algorithm. It is found that, in the face of the influence of complex underwater environment factors, improving the practicability of the algorithms in three-dimensional ocean environment such as ocean currents and complex obstacles, thus achieving AUV's efficient obstacle avoidance and energy saving is the focus of path planning algorithm research. Finally, the research trends of AUV path planning algorithm in intelligent development, fusion and multi-AUV collaboration, and the development of AUV remoteness for wide-area detection are prospected.

In view of problems of redundant search point and poor real-time performance when encountering sudden threat existed in application of traditional algorithm in route planning of unmanned aerial vehicle in environment detection for post-disaster underground, a reverse variable weight sparse algorithm was proposed. According to performance constraint of unmanned aerial vehicle and threat model in post-disaster underground, global static route planning is carried out from target point to start point, so as to avoid a large number of invalid searches. Different weight coefficients are set according to mission requirement of unmanned aerial vehicle, so as to obtain route focuses on distance or safety. Only the path covered by sudden threat is corrected by introducing sub-target point strategy, so as to avoid sudden threat effectively in short time. The simulation results show that the algorithm used in route planning can save time of route planning and reduce threat to unmanned aerial vehicle, which can effectively guarantee real-time performance and safety of route planning.

[Objective] The detection of sky polarization modes is a crucial step in practical applications of polarization navigation. To mitigate the effects of adverse weather conditions on the polarization image acquisition system and protect the polarization camera lens from damage and contamination, a robust and efficient sky polarization mode detection test device has been developed. This paper designs an experimental device for sky polarization mode detection. This device, based on a three-lens polarization camera and equipped with a polarization information processing system, enables real-time and accurate acquisition of atmospheric polarization patterns. Unaffected by rain or poor weather, this device can be used to detect sky polarization patterns in challenging conditions, such as rainy days or over the ocean surface. [Methods] In this study, the experimental device is designed considering three aspects: external device construction, hardware configuration, and software design. A universal ring plays an important role in the test device, ensuring that regardless of the device tilt, the shooting direction of the three-eye camera remains aligned with the sky vertex, thus guaranteeing stable data acquisition. The device adopts a three-channel polarization imaging method to capture the polarization information of the sky. Each polarization channel comprises a polarization device and an optical imaging lens system. The polarization devices in the three channels are arranged at specific polarization angles to each other, with the optical axis direction of the polarizer set to 0°, 60°, and 120°. Therefore, only linearly polarized light from these three directions can enter the charge-coupled device (CCD) camera. By analyzing the distribution pattern of sky-polarized light and applying the Stokes vector analysis model, we designed a host computer for the polarization image processing system. Linear polarization images received by the CCD camera from different directions are input into the polarization image processing system to calculate the sky polarization mode diagram of the target region. [Results] Analysis of the sky polarization mode diagram reveals that the angle of polarized elevation is symmetrically distributed with respect to the solar meridian. The direction of the solar meridian depends on the solar azimuth. The degree of polarization (DOP) presents a regular circular distribution around the neutral point, with the DOP gradually increasing from the neutral point. The distribution of the sky polarization pattern is closely related to the sun's altitude angle. As the sun's altitude angle increases from small to large and then decreases, the DOP follows a similar pattern, reaching its minimum value at noon. This suggests that in the morning or evening, when the sun's altitude angle is small, the DOP is large. Conversely, at noon, when the sun's altitude angle is at its peak, the DOP is small. [Conclusions] Our experimental results demonstrate that the sky polarization pattern detection test device designed in this study successfully mitigates the influence of complex changes and adverse weather conditions. It not only protects the hardware system of image acquisition but also enhances the stability of the polarization image acquisition system. With its impressive real-time performance and robustness, this device can accurately detect and simulate the sky polarization mode of the local observation point in real time through the upper computer of the sky polarization detection system. The device remains unaffected by uneven terrain and landforms. Its excellent sealing and water-resistance properties further ensure the stability of its measurement outcomes. The device provides substantial technical support for polarization navigation and remote sensing technology in complex environments. [ABSTRACT FROM AUTHOR]

A new water surface perception system based on ultrasonic phased array imaging technology has been designed, which can achieve the perception and imaging of target objects in the nearshore sea environment. The ultrasonic wave is emitted by the transmitting transducer, and the echo signal is received by the receiving transducer. After being amplified by an amplification circuit, a sliding average filtering algorithm is introduced for data smoothing processing. The ADC acquisition module is used to convert it into a digital signal for time delay estimation. Then, the target object is imaged using a time delay superposition algorithm, and the Canny operator edge detection is performed. Finally, the size and orientation of the target are obtained. The experiment verified the reliability of hardware circuits and the effectiveness of software algorithms, achieving the design goals of low cost, high portability, and high reliability, and meeting the teaching requirements. [ABSTRACT FROM AUTHOR]