In recent years, extensive research has been conducted on pesticide application technology using unmanned aerial spraying systems (UASS) due to their efficiency and ability to overcome terrain obstacles. However, the coupling effect between the operational parameters of UASS and their influence on droplet deposition has not been sufficiently studied. A thorough and methodical analysis is essential to assess the deposition performance and drift risk of UASS. This study evaluated the spraying performance of an electric six-rotor UASS in wheat fields in Zibo between 2021 and 2022, focusing on three operational modes determined by flight speed and flow rate. Furthermore, the individual effects of these two parameters on droplet deposition quality and drift risk were explored. Based on the deposition quality of in-swath droplets and the drift degree after application, the results demonstrate that the optimal comprehensive characteristics of droplet deposition occur at a flight speed of 4.5 m/s, a flow rate of 2.025 L/min, and a spray amount of 1 L/ha. The increase in spray flow rate (2.475 L/min) results in a 3.92-fold enhancement in the deposition rate within the spray area compared with that of group of the minimum spray flow rate (1.575 L/min). A higher flight speed (5.5 m/s) improves the uniformity of droplet deposition, with the coefficient of variation decreases by 25.2% compared with that of the minimum flight speed (3.5 m/s), and this higher flight speed leads to a drift distance of 28.8 m.

To tackle the navigation challenge in dynamic and complex environments, this study designs a fusion planning framework that synergistically integrates enhanced A* algorithm with improved DWA, inspired by the biological dual-layer navigation mechanism of global path planning and local real-time obstacle avoidance. Firstly, the original global path from the conventional A* algorithm is smoothed and length-reduced through a three-stage optimization strategy involving redundant node removal and forward and reverse path relaxation, mimicking the behavioral logic of honeybees and desert ants that eliminate redundant routes to complete foraging and homing with minimal energy consumption. Secondly, an evaluation function integrating dynamic obstacle perception and adaptive weight adjustment is designed for the DWA to enhance the intelligence of local planning, drawing on the adaptive strategy of animals such as antelopes that adjust behavioral priorities according to environmental complexity to balance safety and efficiency. To comprehensively verify the performance of the proposed algorithm, simulation evaluations are performed in various scenarios, including 20 × 20 and 30 × 30 grid maps, with single and dual dynamic obstacles. Results demonstrate that our algorithm outperforms conventional methods in path length, smoothness, and safety. Further physical verification is carried out on a LiDAR-equipped mobile robot (Shenzhen Yuanchuangxing Technology Co., Ltd., Shenzhen, China) based on the ROS platform, confirming that the algorithm can stably achieve static path tracking and real-time obstacle avoidance in real indoor environments. Consequently, the developed hybrid algorithm delivers a viable and robust solution for autonomous mobile robots to navigate safely and efficiently in unpredictable and complex environments.

This study presents an innovative image encryption algorithm anchored on the two-dimensional Logistic chaotic system, designed to address the limitations of traditional methods in terms of speed and security. The algorithm employs pixel shifting and scrambling, coupled with a segmented grayscale value substitution, to achieve a high level of security with strong key sensitivity and robust resistance to both differential and statistical attacks. Experimental validation demonstrates its superiority with fast encryption speeds, as evidenced by an encryption time of 0.2136 seconds for a 256×256 color image, and an entropy close to its ideal value, indicating excellent randomness. This method also exhibits high performance in resisting differential attacks, with Number of Pixels Change Rate (NPCR) and the Unified Average Changing Intensity (UACI) values nearing their optimal expected values. The integration of Quantum Key Distribution (QKD) is suggested for future work to enhance security against quantum computing threats, solidifying the algorithm’s potential as a leading solution for the image encryption.

In dynamic scattering environments, the secure generation and transmission of images are hindered by issues such as signal distortion and data loss, which degrade encrypted image quality and transmission efficiency. This study proposes a multimodal diffusion network integrating hyperchaotic systems and quantization to address secure image transmission challenges in dynamic scattering environments. The proposed method utilizes hyperchaotic mapping for nonlinear encryption and quantization for efficient compression, resulting in a 0.7 improvement in compression ratio and a 50% enhancement in transmission rate while maintaining image fidelity (PSNR = 20.1 dB, SSIM = 0.52 at 5 dB SNR). The adversarially trained network demonstrates robust adaptability against multimodal attacks, showing 18.8%, 7.7%, 19.7%, 26.4%, and 29.3% anti-interference improvements over single-mode methods for Gaussian noise, pulse interference, band loss, simulated/real dynamic scattering attacks, respectively. The framework ensures global stability in time-varying channels through adaptive strategy optimization, effectively balancing structural integrity with encryption efficiency. Experimental findings substantiate its aptitude to mitigate signal distortion and data loss while preserving security-critical features under complex scattering conditions.

In multimodal image restoration, the unclear causal relationships between degradations and insufficient cross-modal fusion lead to loss of high-frequency details and error accumulation. We proposed an innovative causal-driven framework: CDMRF (Causal-driven framework integrating deep residual learning and uncertainty perception). The framework was evaluated on three publicly available datasets: RainCityscapes (containing 1000 pairs of images with rain streaks and motion blur under various urban settings), SOTS (Synthetic objective testing set) and Flickr2K (providing 2000 natural scene images suitable for verifying the robustness of the model under diverse lighting conditions). CDMRF constructs a causal directed acyclic graph (DAG) for cross-modal degradation, uses the do operator to analyze confounding factors, and applies counterfactual reasoning to separate degradation paths such as rain streaks and noise. A deep residual pyramid network extracts high-frequency residual features in the frequency domain, aligning texture details via cross-modal residual attention. A Bayesian neural network quantifies fusion uncertainty, with a risk-sensitive loss function for pixel-level confidence estimation. Experiments show that at a noise standard deviation of 50, CDMRF maintains a Peak Signal-to-Noise Ratio (PSNR) of 23.5 dB, Cross-Modal Structural Similarity Index (CM-SSIM) of 0.77, and Learned Perceptual Image Patch Similarity (LPIPS) of 0.18. Under full modal fusion (including RGB, near-infrared, depth, thermal, and hyperspectral modalities), PSNR reaches 48.5 dB, CM-SSIM is about 0.96, and LPIPS is 0.092, indicating high restoration quality and feature fidelity.