Xiaohui Lu

How to address the challenges in human–machine cooperative control of intelligent vehicle is still an important issue, which include the conflicting objectives between driver and controller, the difficulty in fusing multi-source risk information from both vehicle and driver states, the limited robustness of conventional risk estimation under uncertain interference. To this end, this paper proposes a human–machine collaborative driving collision avoidance control based on Bayesian risk estimation. Firstly, a Bayesian multi-factor driving risk estimation model is designed by integrating time-to-collision, collision distance and driver fatigue state. Simultaneously, the risk level is used as controller weights within the optimization objective of the non-cooperative game-theoretic controller for resolving conflicting objectives between the driver and the controller. Finally, the effectiveness of the proposed method is verified using PreScan and Simulink co-simulation, along with a hardware-in-the-loop driver testing platform. Experimental results show that under single and double-lane-change scenarios, the proposed strategy exhibits strong disturbance resistance in the white noise and fast response, significantly improving collision avoidance performance in complex environments.

Urban traffic is characterized by frequent speed changes, sudden lane incursions, and heterogeneous vehicle interactions, making adaptive cruise control (ACC) one of the most demanding tasks for intelligent vehicles. Existing DRL-based approaches suffer from fixed-weight reward functions, monolithic experience buffers, and uniform sampling, leading to slow convergence and limited scenario adaptability. This paper presents an integrated SAC-based ACC framework that jointly addresses these issues through: (1) a dynamic reward function that adjusts safety, efficiency, and comfort weights in real time; (2) scenario-specific experience classification that stores and replays samples by driving mode; and (3) a stratified sampling strategy combining recent-experience injection with localized prioritization. The system state fuses bird’s-eye-view semantic segmentation features with longitudinal and lateral motion measurements. Validated in CARLA across six scenarios spanning four core ACC modes—cruising, car-following, cut-in, and cut-out—the method achieves a 38.9% reduction in training episodes over baseline SAC. The trained policy maintains following-distance errors ≤ 1 m, speed-tracking errors ≤ 1 m/s, and peak acceleration generally below 3 m/s², demonstrating effective and comfortable control under dynamic urban conditions.

A novel asymmetric structure of piezoelectric stick–slip actuator (PSSA) with chainlike compliant mechanism is designed, which satisfies the requirements of high-precision, long-stroke nano-scale motion control. The combined structure of asymmetric chainlike compliant mechanism with PSSA is used to enhance the compliance. Backward displacement is suppressed by chainlike piezoelectric stick–slip actuator (CPSSA). The CPSSA is designed to enlarge the single-step displacement and to improve the output performance. Subsequently, finite element analysis is employed to compute the parametric design of the structure. Then, the dynamic stiffness matrix analysis of the structure is given. Finally, an experimental system is constructed, and a range of test data are presented. The experimental findings demonstrate that the peak output speed of the actuator is 19.01 mm/s, while the horizontal loading capability and vertical loading capacity are determined to be 3.8 and 2.46 N, and the resolution is 56 nm, respectively. The effectiveness of the chainlike piezoelectric stick–slip actuator (CPSSA) is thereby verified.

An innovative piezoelectric stick-slip actuator is designed, which incorporates the simple motion mode of ‘contraction-relaxation’ of jellyfish umbrella body and the flexible motion advantage of symmetrical configuration. The actuator features a bionic jellyfish-inspired flexure hinge mechanism (BJFHM), which is consisted by a dual drive foot and a bridge-type mechanism. The deformation within the bridge-type mechanism is conveyed to the dual driving feet, allowing the two sliders to operate either independently or in conjunction. Thereby, enabling a greater variety of motion modes and enhancing operational efficiency. The pseudo-rigid body model and finite element analysis (FEA) are presented. Finally, the experimental results generated by the piezoelectric stick-slip actuator are demonstrated. At 350 Hz, the left-right locking force is 0.5–1.5 N, with velocity reaching 16.44of the left and 2.37 mm/s of the right. The resolutions achieved on the left and right are 51 and 56 nm, respectively. The improvements made with the proposed actuator are confirmed by the results.

A novel piezoelectric stick-slip actuator is designed, which incorporates the strong power output of the mid-foot of a grasshopper with excellent stride characteristics. The actuator utilizes a flexible hinge structure inspired by the mid-foot of a bionic grasshopper, which can achieve large load capacity and high stepping efficiency. Finite element analysis (FEA) is employed to complete the parametric design of the structure, and the rationality of the structure is verified by theoretical analysis. Furthermore, dynamical model is established to clarify the working process. Then, static and fatigue analysis of flexible hinge is completed. Experimental findings reveal that the prototype achieves a peak output speed of 17.28 mm/s and can support a maximum load of 1700 g, which is equivalent to 208 times its own weight of 8.14 g, and the stepping efficiency of the actuator is 99.04%. The findings fully validate the scientific validity and application of the mid-foot design concept in bionic grasshopper.

The conflicting requirements of stiffness and flexibility present a key design challenge for compliant mechanisms in piezoelectric stick-slip actuators. To overcome this, we introduce a multi-objective topology optimization scheme that employs a combined objective function, simultaneously minimizing strain energy and maximizing the displacement amplification ratio. The multi-objective function addresses two key aspects, the strain energy minimization term serves to improve the stiffness of the flexible hinges, whereas the displacement amplification ratio maximization term is designed to increase their deformation, thus achieving greater flexibility in the compliant mechanism. To address the disparity in the magnitude of the objective function, a normalization technique is utilized. The sensitivities are determined using the adjoint method, and the optimization challenge is tackled through the method of moving asymptotes. The stiffness and flexibility of the compliant mechanism are verified by finite element analysis, then the prototype is fabricated. The results from the experiments indicate a load capacity of 170 g and an actuator velocity of 16.57 mm s ⁻¹ . The findings validate the efficiency of the proposed multi-objective topology optimization approach.

Detection of traffic participants is one of the core tasks in environmental perception for autonomous driving. However, accurately identifying small sized and occluded objects in complex traffic scenarios remains a pressing challenge. To address these issues, a high precision multi-scale perception architecture named DCD-YOLO is proposed based on YOLOv11n, which integrates multi-branch feature extraction, attention guided enhancement, and task decoupled prediction mechanisms.A Dynamic Fusion Block (DFB) is introduced to partially replace the original C3k2 structure, enhancing feature extraction for small objects and complex backgrounds. In addition, the Content Guided Attention (CGA) module is incorporated in the Neck to emphasize key object regions through the collaborative effects of channel, spatial, and pixel level attentions.Meanwhile, a Dynamic Shared Enhanced Head (DSEH) combines reparameterizable detail enhanced convolutions, shared convolutional weights, and Group Normalization to improve multi-scale object localization and classification performance while maintaining a lightweight architecture.Experimental results demonstrate that DCD-YOLO improves mAP@0.5 by 2.9% on the KITTI dataset. Furthermore, the proposed model is deployed on a real vehicle platform and tested under realistic driving scenarios, which further verifies its deployment feasibility and practical significance for autonomous driving systems.

A novel structure of a piezoelectric stick-slip actuator is proposed, which is based on the moving posture of a monkey. The biomimetic monkey type of piezoelectric stick-slip actuator (BMPSSA) is designed to simulate the limbs and tail of a moving monkey. By using the pseudo-rigid body method, the deformation model of the compliant mechanism is established. Next, the displacement of flexible hinges is calculated by matrix-based compliance modeling. Then, the modal analysis is computed by the finite element method, respectively. Finally, an experimental system is constructed to test the output performance of the BMPSSA. At the excited frequency of 274 Hz, the experimental results show that the maximum output velocity of BMPSSA reached 21.88 mm/s, the maximum step displacement is 79 μm, and the minimum displacement resolution of BMPSSA is 60.04 nm.

This paper presents the development of a new bionic piezoelectric stick–slip actuator (PSSA). This actuator is designed to simulate the human rowing motion. By synergizing a lever amplification structure with a bridge amplification mechanism, the design not only mimics the rowing motion, where the lever amplification structure acts like an oar and the bridge amplification mechanism resembles the force exerted by a person operating the oar, but also suppresses the problem of backward displacement in traditional PSSAs. Finite element analysis and experimental results demonstrate the feasibility of the design, with a peak velocity of 17.48 mm/s and a maximum load capacity of 55 g. Comparative experiments have shown that the rowing-motion piezoelectric actuation mode outperforms the traditional PSSA mode, with a 16.8% increase in peak velocity and a 120% increase in load capacity. The innovative design has significant implications for the development of high-performance actuators.