Changfu Zong

Driving intention recognition and behaviour prediction based on a double-layer hidden Markov model

We propose a model structure with a double-layer hidden Markov model (HMM) to recognise driving intention and predict driving behaviour. The upper-layer multi-dimensional discrete HMM (MDHMM) in the double-layer HMM represents driving intention in a combined working case, constructed according to the driving behaviours in certain single working cases in the lower-layer multi-dimensional Gaussian HMM (MGHMM). The driving behaviours are recognised by manoeuvring the signals of the driver and vehicle state information, and the recognised results are sent to the upper-layer HMM to recognise driving intentions. Also, driving behaviours in the near future are predicted using the likelihood-maximum method. A real-time driving simulator test on the combined working cases showed that the double-layer HMM can recognise driving intention and predict driving behaviour accurately and efficiently. As a result, the model provides the basis for pre-warning and intervention of danger and improving comfort performance.

Driving intention identification and maneuvering behavior prediction of drivers on cornering

An integrated model, which combines Artificial Neural Network (ANN) and Hidden Markov chain Model (HMM), is proposed to identify the driving intention and predict the maneuvering behavior of the driver, which accordingly realizes intelligent driving assistance and improves active safety of the vehicle. Driving simulator tests on straight line and double-lane change are employed to validate the availability of this model. The test results indicate that this model serves accurate identification on driving intention as well as precise prediction on maneuvering behavior of the driver at certain vehicle speed.

Fault tolerant control method for steer-by-wire system

In this study, we describe a model based fault tolerant control method for steer-by-wire system. First of all general redundancy schemes and current works for steer-by-wire fault tolerant control systems are examined. Then the paper focuses on the fault tolerant control methods of key sensors and DC motors for steer-by-wire system. With the aim to provide a steady and reliable fault tolerant control algorithm we synthesize Adaptive Fading Kalman Filter and fault eigenvectors as a framework to establish the fault tolerant control system. The effectiveness of this research is demonstrated by simulation. Finally key innovations of the fault tolerant control method are discussed.

UKF and EKF estimator design based on a nonlinear vehicle model containing UniTire model

Dynamic model based vehicle state variables observer is a step towards economical on-board sensing system. However a complex model always leads to a control system with a poor real-time performance, while a simple model cannot exhibit real characteristics of a vehicle. In order to make an accurate and sententious estimate for yaw rate and side slip angle, an ameliorated 2-DOF bicycle model containing UniTire model is introduced. Then two observers based on Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF) are introduced. And validity of the two algorithms is verified by simulation test and contrast is brought out respectively. The simulation results show that the UKF based observer performs better in accuracy and computing speed.

Research on information fusion algorithm for vehicle speed information and road adhesion property estimation

To achieve the vehicle speed information and tire-road adhesion property is very important to the vehicle active safety control system. And how to utilize the sensors in vehicle to educe the needed information for these systems while ensuring the accuracy and real-time performance has great significance. In this paper, the state estimation theory of information fusion technology is applied to estimate the vehicle speed and tire-road friction coefficient based on a 3-degree-of-freedom vehicle model containing modified Dugoff tire model, which can not only satisfy the accuracy but also explain the relation of tire-road. Due to the nonlinear characteristic of tire-road behavior, the Extended Kalman Filter (EKF) is applied. By comparing with the outputs signals from CarSim, high accuracy as well as effective availability of this algorithm have been demonstrated.

A Steering-By-Wire fault-tolerance control strategy based on multi-dimension Gauss Hidden Markov Model

Sensor fault-tolerance control technology plays a significant role in maintaining the reliability of a Steering-By-Wire system. Based on the multi-dimension Gauss Hidden Markov Model (HMM) algorithm, this paper presents a steering wheel angle sensor fault-tolerance control strategy in the Steering-By-Wire system using hardware redundancy technique. The impact of this strategy is examined through HIL (Hardware-in-Loop) experiments and test results demonstrate that the strategy proposed and employed here can effectively improve the performance of a Steering-By-Wire vehicle in terms of increasing its reliability and strengthening its security against sensor failures.

Investigations on cornering control algorithm design and road feeling optimization for a Steer-By-Wire vehicle

To improve the handling performance of a Steer-By-Wire (SBW) vehicle, a series of cornering control algorithms are proposed in this paper. Firstly, an algorithm for enhancing the maneuvering in steady-state cornering is presented. Subjective evaluations are then carried based on a driving simulator to demonstrate its superiority against a conventional vehicle with a fixed steering ratio. On this basis, two categories of control algorithms for dynamically correcting and compensating the transient state cornering responses of vehicles are added on. Simulator tests again are conducted to make comprehensive investigations. Test results indicate that the control algorithms for transient state cornering could effectively improve the handling performances of a SBW vehicle, in terms of reducing the work load of drivers, enhancing the track-holding ability and improving steering response properties. Thirdly, optimization approaches on road feeling have been devised and simulated based on a modified steering effort preference function of Chinese drivers. Test results revealed have declared its validation.

Research on algorithm of stability control for tractor semi-trailer

This paper introduces the differential-braking algorithm to control the stability of tractor semi-trailer. First, it analyses the theory of differential-braking control and confirm the control strategy that follows the ideal yaw rate of 2-DOF model. From this point of view, a PID controller has been designed. The result of the co-simulation associated with MATLAB and ADAMS shows that the differential-braking algorithm is adequately effective in the stability control for the tractor semi-trailer.

Research on Torque Ratio Based on the Steering Wheel Torque Characteristic for Steer-by-Wire System

Steer-by-wire system can improve the performance of vehicle handling stability. Removing the mechanical linkages between the front wheels and the steering wheel leads to a key technique of force feedback for steer-by-wire system. In view of the characteristic of variable torque transmission ratio for steer-by-wire system, this paper proposes a method for designing torque ratio based on the steering wheel torque characteristic for steer-by-wire system. It converts the torque ratio design into equivalent assist torque design by analyzing their relationship. It achieves the torque ratio design at different conditions based on the negative equivalent assist torque characteristic curve. Simulations and vehicle experiments are conducted by the proposed method, and the results show that the design goal has been achieved and the steering wheel torque characteristic obtained is very similar to that of the reference car.

Simulation analysis of stability control strategy for heavy single tractor

A 2DOF reference model of heavy single tractor was set up using the yaw rate and the lateral acceleration as control variables based on analysis of tractor instability mechanism. The reference model of tractor was simulation controlled using allocation strategy of braking moment and control strategy of slip ratio with PD algorithm for typical fishhook steering in Trucksim. The simulation results show that this control method is correct and feasible, the control target that enhance tractor stability has been realized.