1 February 2019 in Nonlinear Analysis: Hybrid Systems
6 December 2017 in IEEE Transactions on Vehicular Technology
This paper investigates the integrated optimization of internal powertrain energy management and external vehicles coordination for multiple autonomous and connected hybrid electric vehicles (HEVs). Most existing research on hybrid vehicles powertrain energy management only considers a single vehicle scenario, and is independent from vehicle coordination control. With the advancement of connected and autonomous vehicles, optimizing both the vehicle level coordination and powertrain level power management together can bring in substantial benefits in the overall energy efficiency. In this paper, to enable such an integrated optimization scheme, first, the models of the external vehicle level dynamics and internal powertrain dynamics are built for each vehicle, respectively. Second, by integrating the dynamics of these two interactive levels together, an augmented system is formed. Then the HEVs energy management problem and the decentralized vehicle coordination for a set of HEVs are addressed in a unified framework using Dynamic Programming (DP). The results show that, as a result of the integrated optimization, the vehicles are able to operate close to the most fuel efficient region, and the battery can avoid being depleted or over charged while maintaining same state of charge (SOC) at the beginning and the end. Moreover, 17:9% less fuel consumption can be achieved for the driving cycle considered, compared with the separated optimization method.
26 November 2024 in IEEE Transactions on Intelligent Transportation Systems
In this paper, we investigate the selection of time headway to ensure robust string stability in connected and autonomous vehicle platoons in the presence of signal noise in Vehicle-to-Vehicle (V2V) communication. In particular, we consider the effect of noise in communicated vehicle acceleration from the predecessor vehicle to the follower vehicle on the selection of the time headway in predecessor-follower type vehicle platooning with a Constant Time Headway Policy (CTHP). Employing a CTHP based control law for each vehicle that utilizes onboard sensors for measurement of position and velocity of the predecessor vehicle and wireless communication network for obtaining the acceleration of the predecessor vehicle, we investigate how the implementable time headway is affected by communicated signal noise. We derive constraints on the CTHP controller gains for predecessor acceleration, velocity error and spacing error and a lower bound on the time headway which will ensure robust string stability of the platoon against signal noise. We perform comparative numerical simulations on an example to illustrate the main results.
14 January 2020 in IEEE Transactions on Control of Network Systems
8 October 2021 in IEEE Signal Processing Letters
Person re-identification (re-ID) aims to identify the same persons’ images across different cameras by deploying a trained re-ID model. Nevertheless, the domain bias between different datasets can lead to considerably low identification accuracy when directly applying a trained re-ID model from one dataset to another. Therefore, in order to develop a kind of more transferable re-ID methods with strong robustness, this letter proposes a fast momentum contrast learning framework, which consists of a teacher encoder and a fast momentum encoder, from a reinforcing visual representation learning point of view. The fast momentum encoder produced by the teacher encoder is employed to build a dynamic dictionary and then design the contrast learning loss function together with the teacher encoder. In addition, both the individual loss and the clustering loss are adopted for obtaining the optimal performances in unsupervised person re-ID models. The proposed framework achieves one stage of end-to-end training and experimental results demonstrate that our identification accuracy outperforms existing advanced methods. The code of this work is available at https://github.com/binquanwang/FMCL_UDA.
11 April 2018 in IEEE Transactions on Circuits and Systems II: Express Briefs
A broad range of practical systems need to perform periodic tasks, such as the mechanical hard-drive read-write arm system, the robotic arm, certain circuitry systems, etc. Based on the internal model principle (IMP), repetitive control (RC) is proved to be effective for periodic signals/disturbances tracking/rejection. However, existing methods on RC mostly require constant update of the control input, regardless of whether it is needed or not, which results in much usage of unnecessary communication resources and the wear of the actuators. To overcome this drawback, in this paper, the event-triggered strategy is for the first time considered in the repetitive control system to reduce the update frequency of the control input. A novel event-triggering condition, which takes into account both the current and last periods’ error information, is embedded in a modified repetitive control system. Thus, the control input will not be updated until the designed error signal exceeds a predefined threshold. A continuous-discrete two-dimensional (2D) model is adopted to model and analyze the event-triggered modified repetitive control system (ETMRCS). The performance of the developed control scheme is then verified through the simulation results.