Search

The traditional articulated steering vehicles rely on the hydraulic steering system and segmented body structure to achieve vehicle steering. The folding steering process exhibits strong hysteresis, which reduces the steering maneuverability and also impacts the autonomous driving performance. This paper proposes a novel control method of folding steering based on differential-hydraulic frequency division cooperation for the redundant drive characteristic of distributed-drive articulated steering vehicles. The main contributions of this paper are as follows: 1) By introducing the speed term into the folding angle control, a position-velocity double closed-loop structure is employed to determine the steering demand torque and achieve the variable steering ratio and accurate control of the folding angle. 2) According to the frequency-domain characteristics of the hydraulic and differential steering systems, a first-order low-pass filtering algorithm is utilized to decompose the steering demand torque into high-frequency and low-frequency components. The high-frequency torque is borne by the differential system, while the low-frequency torque is borne by the hydraulic system, thereby addressing the challenge of cooperative control of multiple actuators. 3) An experimental platform for the distributed-drive articulated steering vehicle is constructed, and the credibility of the dynamic model is experimentally validated. 4) The manipulation and path-tracking performances of the proposed folding steering control method are analyzed. The results indicate that, compared to the hydraulic steering system, the average folding angle rise time and phase delay are reduced by 19.8% and 44.4%, respectively, significantly improving the follow-up performance of folding steering. Furthermore, the average lateral position error and heading angle error in path tracking control are reduced by 38.1% and 27.5%, respectively, which contribute to the electrification and intelligent development of articulated steering vehicles.

Hybrid electric vehicles (HEVs) are considered the most practical option for reducing emissions and fuel consumption of conventionally powered vehicles. Energy management strategies (EMSs) are the core technology of HEVs because of decreasing the cost of the system and limiting its negative effects. Equivalent consumption minimization strategy (ECMS) can achieve instantaneous optimal control and has attracted attention in recent years. In this study, an adaptive equivalent consumption minimization strategy (A-ECMS) based on driving cycle recognition is constructed for a parallel HEV. First, select the standard driving cycle and analyze its characteristic parameters. And then training learning vector quantization (LVQ) neural network-based driving cycle recognizer to achieve an average of 98% accuracy. At last, the optimal equivalent factor (EF) is selected for ECMS by recognizing the current driving cycle. It is jointly simulated and analyzed by AVL CRUISE and MATLAB/Simulink software under NEDC and CHTC-LT driving cycle. The results show that compared with the logic-based EMS, in the NEDC driving cycle the 100 km fuel consumption of A-ECMS decreases by 3.8%, and the battery state of charge (SOC) increases by 1.1%. In the CHTC-LT driving cycle, fuel economy improves by 3.6%, proving the superiority of the A-ECMS.

Recently, the synchronous motion planning (SMP) has been viewed as a crucial issue in dual-arm robot systems. In this study, a new SMP scheme based on the Jacobian pseudoinverse formulation is presented and investigated for dual-arm robot systems. Such a scheme is established by discretizing the existing SMP scheme via a special difference formula. The presented discrete-time SMP (DTSMP) scheme is then theoretically analyzed to possess the biquadrate pattern in the end-effector planning error of each arm, thereby guaranteeing the synchronous planning precision of dual-arm robot systems. Simulation results under the system composed of two four-link planar robot arms are provided to further substantiate the effective and superior performance of the presented DTSMP scheme.

Many control problems require solving the algebraic Riccati equation (ARE). Previous studies have focused more on solving the time-invariant ARE than on solving the time-varying ARE (TVARE). This paper proposes a typical recurrent neural network called zeroing neural network (ZNN) to determine the solution of TVARE. Specifically, the ZNN model, which is formulated as an implicit dynamic equation, is developed by defining an indefinite error function and using an exponential decay formula. Then, such a model is theoretically analyzed and proven to be effective in solving the TVARE. Computer simulation results with two examples also validate the efficacy of the proposed ZNN model.