The global path planner is an important part of the navigation system for autonomous differential drive mobile robots (DDMRs). Aiming at the problems such as long calculation time, large number of search nodes, and poor smoothness of path when A* is applied to global path planning, this study proposes an improved bidirectional search Gaussian-A* (BSGA*) algorithm. First, the Gaussian function is introduced to realize the dynamic weighting of the heuristic function, which reduces the calculation time. Secondly, the bidirectional search (BS) structure is adopted to solve the problem of nodes’ repeated search when there are large obstacles between the starting point and the target point. Finally, a multi-layer turning point filter strategy is proposed to further smooth the path. In order to verify the performance of the improved BSGA* algorithm, experiments are carried out in simulation environments with the size of 15 × 15 and 30 × 30, respectively, and compared with the five common global path planning algorithms including ant colony optimization (ACO), D* lite algorithm, and genetic algorithm (GA). The results show that the improved BSGA* algorithm has the lowest calculation time and generates the shortest and smoothest path in the same environment. Finally, the program of the improved BSGA* algorithm is embedded into the LEO ROS mobile robot and two different real environments were built for experimental verification. By comparing with the A* algorithm, Dijkstra algorithm, ACO, D* lite algorithm, and GA, the results show that the improved BSGA* algorithm not only outperforms the above five algorithms in terms of calculation time, length, and total turning angle of the generated paths, but also consumes the least time when DDMR drives along the generated paths.

Nonlinear friction in a conventional drive feed system feeding at low speed is a main factor that contributes to feed drive complexity. A novel two-axis differential micro-feed system is developed in this study to overcome the accuracy limitation of conventional drive feed system. Instead of the screw-rotating-type ball screw adopted in conventional drive feed system, the transmission component of the proposed two-axis differential micro-feed system is a nut-rotating-type ball screw. In this setup, not only the screw but also the nut is driven by a servo motor. By superposing the two high-speed rotary motions (motor–drive–screw and motor–drive–nut) with an equivalent high velocity and the same rotating direction through the novel transmission mechanism, the nonlinear disturbance from the ball screw can be reduced significantly. In addition, both the axes can avoid the creeping transition zone when the table makes a zero-velocity crossing. Note that the motor load switches when the feeding direction of the table is changed, and the nonlinear friction of the table needs to be compensated. Based on this observation, we further present a cross-coupled intelligent second-order sliding mode control that includes a cross-coupled technology, a second-order sliding mode control, a wavelet fuzzy neural network estimator for the friction, and a motor load switch to improve system performance. The proposed cross-coupled intelligent second-order sliding mode control architecture is deployed on a two-axis differential micro-feed system, where numerical simulations and experiments demonstrate excellent tracking performance and friction compensation capability, achieving position tracking error reduced by 45% compared with conventional drive feed system.

The packing cotton picker has become the mainstream cotton-picking equipment due to its advantages of integrated picking and packing and non-stop unloading. Precision packing is an inevitable development trend, and accurate detection of the cotton bale's dimension during the packing process is the first step toward achieving this. Therefore, this paper proposes a geometry-based method to determine the dimension of cotton bale in packing cotton picker. By establishing a mathematical model of the packing process, the relationship between the rotation angle of the rocker arm component and the cotton bale dimension is derived. The reliability and rationality of this method are preliminarily verified through a calculation example. Since the program fails when the rotation angle of the rocker arm component exceeds 55°, the elastic deformation of the packing belt is incorporated for theoretical correction. Finally, the method's accuracy is further validated through field tests, where the absolute error in cotton bale diameter detection is controlled within 50 mm, and the relative error is kept within 2%. This paper lays the foundation for the packing cotton picker to achieve precision packing.

In factory workshops, wearing safety helmets is vital for worker safety. However, current deep learning-based detection methods are often hindered by large model parameters and high computational demands, limiting their deployment in resource-constrained settings. This article introduces YOLO-AFL, a novel lightweight model designed to solve these problems. The algorithm introduces several key optimizations to improve performance without increasing computational load. Firstly, the K-Means++ algorithm is applied during the anchor box preprocessing stage, along with a new distance metric (1 − AIoU), which enhances anchor box size estimation and boosts performance without additional overhead. Secondly, by introducing a lightweight PConv operation into the C3 module, the complexity of the model is significantly reduced. Finally, a dual attention network (LDA-GC) is designed to compensate for any accuracy loss caused by the model’s simplifications. Experimental results on a custom dataset show that the proposed algorithm achieves an mAP50 of 94.1%. Compared to the baseline model, it reduces the number of parameters by 19.1% and decreases computational complexity by 16.9%, demonstrating its superior performance and efficiency in safety helmet wearing detection.