Unmanned aerial vehicles (UAVs) are commonly employed in pursuit and rescue missions, where the target’s trajectory is unknown. Traditional methods, such as evolutionary algorithms and ant colony optimization, can generate a search route in a given scenario. However, when the scene changes, the solution needs to be recalculated. In contrast, more advanced deep reinforcement learning methods can train an agent that can be directly applied to a similar task without recalculation. Nevertheless, there are several challenges when the agent learns how to search for unknown dynamic targets. In this search task, the rewards are random and sparse, which makes learning difficult. In addition, because of the need for the agent to adapt to various scenario settings, interactions required between the agent and the environment are more comparable to typical reinforcement learning tasks. These challenges increase the difficulty of training agents. To address these issues, we propose the OC-MAPPO method, which combines optimal control (OC) and Multi-Agent Proximal Policy Optimization (MAPPO) with GPU parallelization. The optimal control model provides the agent with continuous and stable rewards. Through parallelized models, the agent can interact with the environment and collect data more rapidly. Experimental results demonstrate that the proposed method can help the agent learn faster, and the algorithm demonstrated a 26.97% increase in the success rate compared to genetic algorithms.

Periodic signal detection methods are widely used in applications including human detection and machinery fault diagnosis. Averaging is one of the most powerful filtering techniques for periodic signals extraction. Time domain synchronous average (TSA) and moving average (MA) are the most commonly used average techniques in engineering. TSA has the advantage at periodic signal detection by depressing noises and asynchronous signal components. MA is effective to remove noises while keeping signal periodicity. However, the TSA signal is not periodic as a measurement signal, and signal spectrum resolution degrades seriously; meanwhile, the MA filters out high-frequency signal components of interests. Detection of periodic signal among noises while keeping signal periodicity and high-frequency signal components become a challenge. To address this problem, time-synchronous moving average (TSMA) method is proposed as an improvement on TSA inspired by MA in this paper. Influences of signal overlap and properties of TSMA are investigated. Furthermore, a practical average times optimization method is given for reference. The correctness of theoretical deviations and effectiveness of the proposed method on periodic signal detection are validated using numerical simulations. At last, the proposed method is validated by an application on fault detection of the gearbox.

Background: In the past decade, the number of reported cases of scrub typhus (ST) has increased dramatically in Sichuan Province. We aimed to overview the epidemiological characteristics of ST, identify the variables contributing to the spatial distribution, and estimate the risk areas of ST occurrence.Methods: Daily ST cases reported at the county level from 2006 to 2021 and datasets on environmental and socioeconomic variables were obtained. Joinpoint regression model was utilized to examine the incidence trends and to calculate the annual percentage change. Global spatial autocorrelation analysis was employed to explore the spatial temporal patterns. Then BRT model was employed to identify variables that make sense and predict the risk areas of ST occurrence.Result: It has been reported that there were 6,338 ST cases in Sichuan Province from 2006 to 2021, and the incidence rates continued to rise. Most cases were distributed between June and October each year, peaking in August. During the study period, the cases showed spatial clustering at the county level, mainly in the Panxi area, and then slowly spread to the northwest and northeast. Shrubs, precipitation, farmland and maximum temperature were the primary variables that affected the spatial distribution of this disease. It was estimated that the areas including Liangshan, Panzhihua, Bazhong, and Guangyuan were most at risk of transmission. and there were approximately 32.315 million people living in the areas with potential risk of infection throughout Sichuan.Conclusion: Many counties in Sichuan Province were estimated to be susceptible to ST. Our found in this data-driven study could be used to guide the implementation of targeted prevention and control measures in high-risk areas.