To improve the control performance of a permanent magnet synchronous motor (PMSM) under external disturbances, an improved active disturbance rejection control (IADRC) algorithm is proposed. Since the nonlinear function in the conventional ADRC algorithm is not smooth enough at the breakpoints, which directly affects the control performance, an innovative nonlinear function is proposed to effectively improve the convergence and stability. On this basis, the proposed IADRC is constructed, and comparative simulation results with ADRC and other IADRC show that faster response speed, higher accuracy and stronger robustness are obtained.

Automated kiwi harvesting hinges on the seamless deployment of a detection model and the accurate detection of kiwifruits. However, practical challenges, such as the limited computational resources on harvesting robots and occlusions among fruits, hinder the effectiveness of automated picking. To address these issues, this paper introduces EDT-YOLOv8n, a lightweight and efficient network architecture based on YOLOv8n. The proposed model integrates the Effective Mobile Inverted Bottleneck Convolution (EMBC) module to replace the C2f modules, mitigating the channel information loss and bolstering generalization. Additionally, the DySample upsampler, an ultra-lightweight and effective dynamic upsampler, improves feature extraction and resource efficiency when compared to traditional nearest-neighbor upsampling. Furthermore, a novel Task Align Dynamic Detection Head (TADDH) is implemented, incorporating group normalization for a more efficient convolutional structure and optimizing the alignment between the classification and localization tasks. The experimental results reveal that the proposed EDT-YOLOv8n model achieves higher precision (86.1%), mAP0.5 (91.5%), and mAP0.5-0.95 (65.9%), while reducing the number of parameters, the number of floating-point operations, and the model size by 15.5%, 12.4%, and 15.0%, respectively. These improvements demonstrate the model’s effectiveness and efficiency in supporting kiwifruit localization and automated harvesting tasks.

To help understand the flow field mechanisms behind the entire process of a flapping-wing rotor (FWR) from startup to dynamic equilibrium of rotary motion, this study develops a dynamic fluid–body interaction (DFBI) model to predict the aerodynamics of the FWR, wherein the DFBI model is used to accurately realize the wing's motions within the simulation. The computational fluid dynamics model in this work is rigorously validated through grid independence studies, time-step independence studies, model independence studies, and comparison with prior data. An experimental lift measurement platform is also constructed. Based on these tools, this study investigates the coupling mechanism between transient rotary speed and flow field and the effect of reduction ratio on the FWR's aerodynamic performance. Key findings indicate that rotary motion significantly suppresses negative lift during the upstroke. In addition, releasing the rotary degree of freedom alters vortex dynamics, shifting the leading-edge vortex core and affecting its structure during downstroke and upstroke. A reduction ratio of 28 is identified as optimal, yielding approximately 20% higher average lift compared to a reduction ratio of 24.5. Overall, this work facilitates the understanding for the FWR's dynamic equilibrium of rotary motion and offers valuable design guidance for further enhancing the lift performance of the FWR.

Flexible actuators are often used to grab fragile and vulnerable items in many fields. Compared with rigid actuators, they are safer to interact with humans and can better adapt to unstructured environments. Therefore, to address the poor adaptability of rigid actuators, an optimized pneumatic grid flexible actuator (PGFA) is proposed and analyzed. First, based on the bending principle of the PGFA’s chamber volume change, an improved bending mathematical model for PGFA was established that relates the input air pressure to the bending angle. Next, using this model, an improved Zebra optimization algorithm was proposed to optimize the PGFA structural parameters. Finally, the PGFA was fabricated using a mold-casting method based on 3D printing technology, through finite element simulation and physical testing, the impact of different structural parameter sizes on the bending performance was analyzed, and the bending angles and tip forces of PGFAs with different sizes were compared. The results demonstrate that compared with other pneumatic mesh flexible actuators, the optimized PGFA proposed in this paper significantly improves both the bending angle and tip force, under the same pressure of 40 kPa and 5 knuckles, the bending angle of PGFA optimized in this paper can reach 86.2°, and the tip force can reach 1.4 N. Besides, this study experimentally demonstrates that structural parameter optimization significantly enhances PGFA performance, offering a quantitative benchmark for flexible-actuator theory.

Current passive anti-rollover systems exhibit inadequate adaptability to complex operational environments. Additionally, due to unidentified critical factors driving rollover incidents during landing, the design of active anti-tipping systems for airdrop remains constrained. Given the foregoing circumstances, this paper divides the landing impact process of the vehicle into the airbag cushioning stage and the rigid collision stage. In the airbag cushioning stage, a vertical impact test bench and a fluid–structure interaction (FSI) model is built up to obtain the terminal impact velocity when the airbag’s touching down speed is set as around 8 m/s. An oblique impact test bench and a dynamic model are proposed to investigate the influence of terminal sideslip angles and impact velocities on the vehicle’s roll/pitch stability during the rigid collision phase. Experimental and numerical analyses reveal that the peak overload during the airbag cushioning stage reaches approximately 11 g while the terminal impact velocity in this stage is around 2 m/s. In the rigid collision stage, higher initial descent velocities amplify the peak roll angles and significantly compromise the roll stability. Notably, adjusting the terminal sideslip angle from 90° to 0°/180° triples the critical horizontal velocity threshold from 5.3 m/s to 14.7 m/s which markedly enhances the vehicle’s stability. To address this, an active sideslip angle control system activated at a 250 m altitude is developed to align the vehicle’s horizontal velocity vector with its longitudinal axis to nearly 0°/180° and thus improves the roll/pitch stability. This study establishes a technical foundation for the design of a highly reliable anti-rollover device for the airdrop vehicle.