4 October 2024 in IEEE Transactions on Intelligent Transportation Systems
With the booming adoption of Electric Vehicles (EVs) globally, the need for reliable and resilient EV Charging Monitoring (EVCM) systems has become crucial. A major challenge in real-time EVCM is the handling of missing data caused by unexpected events, which can impair both real-time monitoring and its downstream applications. To address this vital yet underexplored issue, we propose a curriculum guided multi-feature fusion transformer (CurriFusFormer) learning framework – a novel approach designed to enhance the resilience of EVCM systems against real-time information omissions. Our framework integrates curriculum learning with a multi-feature fusion transformer model, capable of handling various patterns and rates of missing data, ranging from random to block omissions. This innovative approach leverages spatial, temporal, and static features to generate accurate real-time estimations for missing values in diverse scenarios. Extensive experiments on a real-world EVCM dataset demonstrate that CurriFusFormer can perform well with $R^2$ ranging from 0.92 to 0.83 given the rising missing rate from 30-90%, outperforming seven popular and state-of-the-art methods, especially in scenarios with high missing rates and complex patterns, such as, at 90% missing rate, kNN ( $R^2$ $=$ 0.65), XGBoost ( $R^2$ $=$ 0.78), BRITS ( $R^2$ $=$ 0.79), TFT ( $R^2$ $=$ 0.80), and GRIN ( $R^2$ $=$ 0.82). All results suggest that the proposed framework could be a promising solution for developing future resilient EVCM networks.
11 July 2024 in IEEE Transactions on Intelligent Vehicles
The perception system based on roadside LiDAR is an important component for autonomous driving. However, previous studies optimizing the deployment configuration of roadside LiDAR systems were constrained by LiDAR designs that failed to provide uniform coverage of vehicle point clouds within the perception area. Our objective is to enhance the perception capability of roadside sensing systems by considering both the inherent properties of LiDAR (e.g., beam distribution) and its relationship to roadside infrastructure, particularly in terms of placement height. Firstly, we proposed an analytical model to establish a quantitative relationship between LiDAR configurations and the distribution of vehicle point clouds. We conducted a field test to generate real vehicle point clouds and validate the proposed analytical model using three types of LiDARs. Secondly, we optimized the roadside LiDAR configurations to maximize the perception entropy of vehicle point clouds. We designed a particle swarm optimization algorithm using initial configurations generated through the bisection method as prior knowledge. The use of a particle swarm optimization algorithm narrowed down the search area, thereby expediting the optimization process. A set of simulation-based experiments was conducted using the NGSIM I-80 dataset. Compared to three alternative configurations, the proposed joint optimization model showed the most significant improvement for the 16-beam and 32-beam LiDAR models, with detection Recall increasing by 65% and 63%, respectively. These improvements are 4.8 times and 3 times higher than those of the base configuration. This study provides an effective method for LiDAR manufacturers and traffic professionals to optimize roadside LiDAR systems, thereby enhancing vehicle perception capabilities.
29 October 2018 in IET Intelligent Transport Systems
Mandatory lane change (MLC) is a critical step in formulating global route of an autonomous vehicle on the urban road network. Improper MLC decisions on arterial roads could jeopardise efficiency (travel cost) and reliability (possibility of failed lane change). However, the existing research studies seldom investigate the optimal MLC decision at the planning level to maximise the reliability‐based driving efficiency. This research aims at addressing two core strategic decision variables (MLC decision point on the road and maximum waiting time before giving up MLC). A series of simulation experiments for various scenarios are conducted for an arterial road to reveal their relationship. The results indicate that both identified decision variables inherently affect travel time spent on this road and the rate of failed MLCs, and a trade‐off exists between arterial travel time and the rate of failed MLCs. Based on the simulation analysis, an analytical lane‐level link performance (LLP) function is formulated to assess the impacts of MLC decisions on driving efficiency. The analysis validates that the optimal MLC strategic decision (MLC decision position and maximum waiting time) can be determined by maximising LLP function. It is promising to apply the proposed LLP function in lane‐level routing algorithm in the future.
12 March 2020 in IEEE Systems Journal
Recent advances in vehicle automation technologies have opened up new perspectives for motorway traffic condition monitoring and macroscopic controls. When sufficient automated vehicles (AVs) are present, they can provide wide-ranging and spatiotemporally detailed traffic information. Furthermore, AVs can be programmed to comply with traffic laws and optimized to smooth traffic flows. Thus, deploying AVs are expected to address the issues encountered by the existing variable speed limit (VSL) systems. However, the appealing scenario that 90% or more AVs are on motorways is seen as a long-term goal. Therefore, this article explores the utilization of AV technologies in VSL systems under mixed traffic conditions where AVs coexist with manually driven vehicles. In this article, a VSL system using AVs as a source of data is presented. More specifically, an extended Kalman filter-based data assimilation method is proposed to estimate system variables (i.e., density, speed, critical density, and compliance rate) from the collected AV data. Following this, a model predictive control scheme is adopted to solve the optimal AV-data-based VSL control problem on the basis of estimated system variables. Finally, the efficiency of the proposed system is verified against a real motorway section in New Zealand.
11 July 2024 in IET Intelligent Transport Systems
The fundamental diagram (FD) of traffic flow can effectively characterize the macroscopic characteristics of traffic flow and provide a theoretical foundation for traffic planning and control. The rapid development of connected vehicles (CVs) has led to changes in traffic flow characteristics. However, research on the FD of traffic flow involving CVs and non‐connected vehicles (NCVs) is still in its early stages. Most FDs do not well characterize the motion behaviour of different vehicles, nor do they study the interaction between mixed vehicles. Therefore, in this study, the FD of mixed traffic flows (i.e. with CVs and NCVs) was constructed within a unified framework. First, the car‐following behaviours of CVs and NCVs were modelled based on risk potential field theory. Subsequently, the FD of mixed traffic flows was derived based on the relationship between car‐following behaviour and the macroscopic traffic flow under steady‐state conditions. To validate the model, rigorous verifications were conducted via numerical experiments using the Monte Carlo method. The results indicate significant agreement between the scatter plots obtained from the experiments and the theoretical curves for different penetration rates. The proposed FD has a unified framework and a more rigorous mathematical structure.
1 March 2024 in Journal of Transportation Engineering, Part A: Systems