In smart-city deployments, vast spatiotemporal trajectory data are generated by consumer-electronics sensors, such as in-vehicle devices. Predicting the future movements of traffic participants in such complex environments is a fundamental task in autonomous driving. In this study, we proposed a trajectory prediction strategy based on attention mechanism and self-supervised learning. First, the agent and map features are extracted, and a set of searched candidate paths for the target agent is constructed based on prior knowledge. Leveraging these candidate paths, a feature selection mechanism is introduced to enable the model to adaptively select relevant feature vectors. This enhances the attention mechanism’s ability to model agent interactions. Furthermore, self-supervised learning is utilized to design auxiliary tasks, including path classification and agent interaction strength estimation, which further improve trajectory prediction performance. Extensive experiments on the nuScenes dataset demonstrate that the prediction model is competitive with the state-of-the-art models, reducing minADE by up to 32.6% and minFDE by 40.8% compared to the latest DICE model, while maintaining real-time inference with an average runtime of 14.9 ms per frame for 8 agents. Ablation studies indicate that each component is essential to ensure prediction performance. Finally, closed-loop Vehicle-in-the-Loop experiments validate the model’s ability to generate meaningful multimodal predictions and effectively integrate them into downstream planning tasks, the prediction-aware planner yields enhanced safety and comfort over existing baselines, and its low computational cost highlights strong potential for deployment on consumer electronics platforms, laying the groundwork for intelligent spatiotemporal analytics in smart-city traffic management.

Platooning is a remarkable research direction in connected electric vehicles for improving energy efficiency. This study proposes an energy-efficient predictive and reactive control strategy of connected electric vehicles platoon with string stability based on variable time headway. Firstly, a trajectory prediction method is designed to predict future motion of the adjacent preceding vehicle of a platoon, according to the information of downstream vehicles. Combining the predictive trajectory of the adjacent preceding vehicle with model predictive control (MPC), the platoon leader executes energy-saving velocity optimization to achieve macro-level energy efficiency for the entire platoon. Secondly, a reactive controller is proposed to further improve energy-saving potential and guarantee stability. It is committed to achieving dual energy-saving effects, and guaranteeing the string stability of the platoon, by designing an enhanced variable time headway policy and a distributed cascaded proportional integral derivative control algorithm. Subsequently, a real dataset is employed to verify the effectiveness of the proposed control strategy. The results indicate the excellent performance of our proposed control strategy in three traffic states. Finally, the real-time requirement of the control strategy is verified by hardware-in-the-loop testing.

Uncertainties, including parameter variation and unknown external disturbances, are the major concerns in the development of a controller, which may lead to vehicle instability when driven in a critical state under extreme conditions. This article proposes an enhanced control strategy considering uncertain parameter (tire stiffness) and external disturbance (road-tire friction coefficient). Initially, the general polynomial chaos (gPC) is applied to model the randomness of parameters. Subsequently, a stability margin that accounts for stochastic uncertainties, including variations in road conditions and tire characteristics, is established. This foundation enables the precise definition and implementation of a control strategy. A reference generator with a variable desired yaw rate is developed to modify the driving characteristics and facilitate personalized handling-oriented requirements. Furthermore, a second-order sliding mode controller determines the external yaw moment with robustness and simple deployment. Ultimately, the experimental results demonstrate the effectiveness in improving handling and stability, which is robust to road friction uncertainty while reducing the limit load on the motor. Compared to the benchmark method, the maximum sideslip angle is reduced by 8.6%–9.3%, and the response time lag is enhanced by 60%.