SUMMARY: This study proposes and investigates a new type of bi-criteria minimization (BCM) for the motion planning and control of redundant robot manipulators to address the discontinuity problem in the infinity-norm acceleration minimization (INAM) scheme and to guarantee the final joint velocity of motion to be approximate to zero. This new type is based on the combination of minimum weighted velocity norm (MWVN) and INAM criteria, and thus is called the MWVN–INAM–BCM scheme. In formulating such a scheme, joint-angle, joint-velocity, and joint-acceleration limits are incorporated. The proposed MWVN–INAM–BCM scheme is reformulated as a quadratic programming problem solved at the joint-acceleration level. Simulation results based on the PUMA560 robot manipulator validate the efficacy and applicability of the proposed MWVN–INAM–BCM scheme in robotic redundancy resolution. In addition, the physical realizability of the proposed scheme is verified in practical application based on a six-link planar robot manipulator.

In order to solve the problem of how to perform path planning for AUVs with multiple obstacles in a 3D underwater environment, this paper proposes a six-direction search scheme based on neural networks. In known environments with stationary obstacles, the obstacle energy is constructed based on a neural network and the path energy is introduced to avoid a too-long path being generated. Based on the weighted total energy of obstacle energy and path energy, a six-direction search scheme is designed here for path planning. To improve the efficiency of the six-direction search algorithm, two optimization methods are employed to reduce the number of iterations and total path search time. The first method involves adjusting the search step length dynamically, which helps to decrease the number of iterations needed for path planning. The second method involves reducing the number of path nodes, which can not only decrease the search time but also avoid premature convergence. By implementing these optimization methods, the performance of the six-direction search algorithm is enhanced in favor of path planning with multiple underwater obstacles reasonably. The simulation results validate the effectiveness and efficiency of the six-direction search scheme.

Summary: This paper presents a neural network-based four-direction search scheme of path planning for mobile agents, given a known environmental map with stationary obstacles. Firstly, the map collision energy is modeled for all the obstacles based on neural network. Secondly, for the shorted path-search purpose, the path energy is considered. Thirdly, to decrease the path-search time, a variable step-length is designed with respect to collision energy of the previous iteration path. Simulation results demonstrate that the variable step-length is effective and can decrease the iteration time substantially. Lastly, experimental results show that the mobile agent tracks the generated path well. Both the simulation and experiment results substantiate the feasibility and realizability of the presented scheme.

As a critical energy component in electric vehicles, energy storage systems, and other applications, the accurate estimation of the State of Health (SOH) of lithium-ion batteries is crucial for performance optimization and safety assurance. To this end, this paper proposes a hybrid model named Transformer–KAN, which integrates Transformer architecture with Kolmogorov–Arnold Networks (KANs) for precise SOH estimation of lithium-ion batteries. Initially, five health features (HF1–HF5) strongly correlated with SOH degradation are extracted from the historical charge–discharge data, including constant-voltage charging duration, constant-voltage charging area, constant-current discharging area, temperature peak time, and incremental capacity curve peak. The effectiveness of these features is systematically validated through Pearson correlation analysis. The proposed Transformer–KAN model employs a Transformer encoder to capture long-term dependencies within temporal sequences, while the incorporated KAN enhances the model’s nonlinear mapping capability and intrinsic interpretability. Experimental validation conducted on the NASA lithium-ion battery dataset demonstrates that the proposed model outperforms comparative baseline models, including CNN–LSTM, Transformer, and KAN, in terms of both RMSE and MAE metrics. The results indicate that the Transformer–KAN model achieves superior estimation accuracy while exhibiting enhanced generalization capabilities across different battery instances, indicating its strong potential for practical battery management applications.

This paper proposes a four-direction search method for obstacle avoidance of mobile robots, and the collision energy of obstacles is modeled based on the neural network. For comparison and discussion purposes, the different invariant step lengths are tested in the static environment. To take the advantages of the large and small step lengths effectively, a variable step length method is adopted to improve the performance, such as less iteration number and lower total energy. The variable step length method is also applied to the dynamic environment to explore the real-time performance for path planning. Simulation results demonstrate the effectiveness and practicability of the presented scheme.

With the rapid increase in vehicle ownership and increasingly stringent emission regulations, addressing the energy consumption of and emissions from commercial vehicles have become critical challenges. This study introduces a multi-objective cooperative adaptive cruise control (CACC) strategy, designed for intelligent connected commercial vehicle platoons, operating in event-triggered conditions. A hierarchical control framework is utilized: the upper layer handles reference speed planning based on vehicle dynamics and constraints, while the lower layer uses distributed model predictive control (DMPC) to manage vehicle following. DMPC is chosen for its ability to manage distributed platoons by enabling vehicles to make local decisions, while maintaining system-wide coordination. Additionally, adaptive particle swarm optimization (APSO) is employed during the optimization process to solve the optimal problem efficiently. APSO is employed for its computational efficiency and adaptability, ensuring quick convergence to optimal solutions with reduced overheads. An event-triggering mechanism is integrated to further reduce the computational demands. The simulation results show that the proposed approach reduces fuel consumption by 8.05% and NO_x emissions by 10.15%, while ensuring stable platoon operation during dynamic driving conditions. The effectiveness of the control strategy is validated through extensive simulations, highlighting superior performance compared to conventional methods.