Search Results

LiDAR and camera fusion have emerged as a promising approach for improving place recognition in robotics and autonomous vehicles. However, most existing approaches often treat sensors separately, overlooking the potential benefits of correlation between them. In this paper, we propose a Cross- Modality Module (CMM) to leverage the potential correlation of LiDAR and camera features for place recognition. Besides, to fully exploit potential of each modality, we propose a Local-Global Fusion Module to supplement global coarse-grained features with local fine-grained features. The experiment results on public datasets demonstrate that our approach effectively improves the average recall by 2.3%, reaching 98.7%, compared with simply stacking of LiDAR and camera.

As an important component of the Intelligent Transportation System (ITS), short-term traffic flow prediction is a key step to assess the traffic situation. It provides suggestions for travellers and helps the administrators manage the traffic effectively. Due to the availability of massive traffic data with various features, the data-driven methods have been applied widely to improve the accuracy of traffic flow prediction. However, few previous studies try to capture the information of traffic flows from multi entry stations to forecast the overall tendency of traffic flow. In this paper, we collect data at a highway exit station in Shanghai, split the data according to originating entry stations and predict the corresponding exit station traffic flow from that of the multi entry stations. Firstly, the original records are collected, preprocessed, aggregated and normalized.

Intelligent Transportation System (ITS) plays an important role in smart city, and accurate short-term traffic flow prediction is a significant part. At present, China’s ITS has developed rapidly, and advanced intelligent transportation systems have been built in major cities, such as Shanghai, Shenzhen and so on. With the promotion of mixed Electronic Toll Collection (ETC) and Manual Toll Collection (MTC) charging systems, the features of the traffic flow data have become richer. Traffic data recorded some information for the vehicles entering and exiting highway toll station including time, location, type, mileage, then we can use historical OD data to do traffic flow prediction, predict the corresponding future exit station traffic flow. Furthermore, due to the deep learning network’s ability to model deep complex non-linear relationship in data, researchers have paid more attention to predict traffic flow using deep learning models in recent years.

Accurate road friction coefficient is crucial for the proper functioning of active chassis control systems. However, road friction coefficient is difficult to be measured directly. Using the available onboard sensors, a model-based Extended Kalman filter (EKF) algorithm is proposed in this paper to estimate road friction coefficient. In the development of estimation algorithm, vehicle motion states such as sideslip angle, yaw rate and vehicle speed are first estimated. Then, road friction coefficient estimator is designed using nonlinear vehicle model together with the pre-estimated vehicle motion states. The proposed estimation algorithm is validated by both simulations and tests on a scaled model vehicle.

In this paper, based on our previously preliminary out-of-plane tire model, a complete out-of-plane flexible tire model is further developed by considering the variation of dimension and parameter values among different slices of the tire model. This tire model is validated via various MSC ADAMS® FTire virtual cleat tests. Especially, the cleat tests with non-zero tire camber angles and non-symmetric cleat shapes, which can better capture the out-of-plane tire properties, are included. By comparing the predicted results of the proposed tire model with FTire for various cleat tests, it shows that the complete out-of-plane flexible ring tire model is better at fully representing the actual tire properties for some complicated cleat testing scenarios.

In this paper, a gain-scheduling optimal control approach is proposed to enhance yaw stability of articulated commercial vehicles through active braking of the proper wheel(s). For this purpose, an optimal feedback control is used to design a family of yaw moment controllers considering a broad range of vehicle velocities. The yaw moment controller is designed such that the instantaneous tractor yaw rate and articulation angle responses are forced to track the target values at each specific vehicle velocity. A gain scheduling mechanism is subsequently constructed via interpolations among the controllers. Furthermore, yaw moments derived from the proposed controller are realized by braking torque distribution among the appropriate wheels. The effectiveness of the proposed yaw stability control scheme is evaluated through software-in-the-loop (SIL) co-simulations involving Matlab/Simulink and TruckSim under lane change maneuvers.

Vehicle jackknifing is generally associated with the loss of yaw stability, and is one of the most common cause of serious traffic accidents involving tractor-semitrailer combinations. In this paper, an active braking control strategy is proposed for jackknifing prevention of a tractor-semitrailer combination on a low friction road. The proposed control strategy is realized via upper-level and lower-level control structures considering braking of both the units. In the upper-level control, the required corrective yaw moments for tractor and semitrailer are generated using a PID controller aiming to reduce errors between the actual yaw rates of tractor-semitrailer and the target yaw rates deduced from a reference model. The corrective yaw moments are achieved through brake torque distribution among the tractor and semitrailer axle wheels in the lower-level control.

A tire may be one of the most critical and complex components in vehicle dynamics and road loads analyses because it serves as the only interface between the road surface and the vehicle. Extensive research and development activities about vehicle dynamics and tire models have been published in the past decades, but it is still not clear about the applications and parameter identification associated with all of these tire models. In this literature review study, various published tire models used for vehicle dynamics and road loads analyses are compared in terms of their modeling approaches, applications and parameters identification process and methodologies. It is hoped that the summary of this literature review work can help clarify and guide the future research and development direction about tire modeling.

This paper presents the implementation of an off-line optimized torque vectoring controller on an electric-drive vehicle with four in-wheel motors for driver assistance and handling performance enhancement. The controller takes vehicle longitudinal, lateral, and yaw acceleration signals as feedback using the concept of state-derivative feedback control. The objective of the controller is to optimally control the vehicle motion according to the driver commands. Reference signals are first calculated using a driver command interpreter to accurately interpret what the driver intends for the vehicle motion. The controller then adjusts the braking/throttle outputs based on discrepancy between the vehicle response and the interpreter command.