Path Preview Research Articles

MPC-based Lateral Motion Control for Autonomous Vehicles Through Serially Cascaded Discretization Method Considering Path Preview

MPC-based Lateral Motion Control for Autonomous Vehicles Through Serially Cascaded Discretization Method Considering Path Preview

Design and experimental evaluation of an efficient MPC-based lateral motion controller considering path preview for autonomous vehicles

Lateral motion control, a core autonomous driving technology, still faces the significant challenge of accurately tracking the reference path under complex and changeable driving maneuvers. In this regard, this study develops an efficient model predictive control (MPC)-based lateral motion controller considering path preview to improve the robustness and computational efficiency in large lateral acceleration or high-speed lateral motion control. As a typical model-based approach, the accuracy of the MPC predictive model substantially affects the controller’s robustness. Thus, to improve the robustness of the lateral motion controller, a tire parameter online adaptive module (TPOAM) is proposed to update the MPC predictive model online to reduce the model mismatch. Unlike other online adaptation methods, the proposed TPOAM does not rely on complex and multi-parameter tire models or look-up tables. Through delay augmentation, the proposed method accurately models the steering system delay and compensates it in the MPC predictive model. In the engineering implementation of the MPC-based lateral motion controller, the long prediction horizon substantially deteriorates the computational efficiency of the receding optimization. Addressing this issue, the preview-follower theory is introduced into the predictive model to make full use of the future path information. The tracking deviation at the corresponding preview point can be obtained by previewing maneuvers similarly to human drivers at each time in the prediction horizon. This tracking deviation is also considered part of the MPC cost function. The total prediction horizon of the lateral motion controller considering path preview can be effectively extended while almost no computing time increment. The designed controller is verified on an autonomous vehicle platform in the low-speed large curvature and high-speed lane changing scenarios. Experimental results show that the proposed controller can effectively improve the robustness and computational efficiency of lateral motion control compared with the typically used MPC-based lateral motion approach.

A Fuzzy Path Preview Algorithm for the Rice Transplanting Robot Navigation System

Path preview is the key to realizing automatic navigation control that can guide the navigational direction of the vehicle trace. The preview point must be regulated dynamically according to the predetermined navigational features of the actual path preview control to obtain the ideal path trace result. Based on the “preview follower theory” combined with the driver’s experience and utilizing the Matlab (The MathWorks, Inc., Natick, Massachusetts, USA) fuzzy control toolbox, a fuzzy algorithm of the preview point has been proposed in the paper. The navigation control system embedded in the path planning and preview control algorithm has been applied in the experiment to trace a predetermined navigational straight way, 90°curve and arc way to operate a rice transplanting robot. The results showed that the lateral DRMS (distance root mean square error) of the straight trace is approximately 0.33 m, and the maximum error is approximately 0.71 m; the lateral DRMS of the 90°curve trace is approximately 0.73 m, and the maximum error is approximately 1.85 m; the lateral DRMS of the arc way trace is approximately 1.03 m, and the maximum error is approximately 2.56 m; and the total mean error is approximately 0.6 m. According to the outcome of the experiment, the relevant algorithm can basically meet the requirements of rice transplanting robot guidance.

Path Tracking Control of a Motorcycle Based on System Identification

This study investigates the roll angle tracking and path tracking controls of a motorcycle. For the control design, a required linear model is obtained from the system identification method. The roll angle is controlled at a specific speed via a simple PID controller to generate input-output data, including steering torque, as well as roll and steering angles. The collected data are then used to identify a one-input two-output linear model by a prediction error identification method using parameterization in a canonical state-space form derived as a Whipple model. With the obtained linear model, full-state feedback (FSF) is designed for roll angle tracking control. Simulations and comparisons with a PID controller show that this FSF is robust against changes in speed as well as external disturbances. On the basis of the roll angle tracking controller, a path tracking controller by path preview is developed with consideration of disturbance rejection. The simulations show the effectiveness of the proposed control scheme.

Integrating GIS and GPS to realise autonomous navigation of farm machinery

Path planning is a key technique to realise autonomous navigation of farm machines. Based on MapObjects, a Geographic Information System (GIS) for autonomous navigation was developed with Visual C++. A path search algorithm used to guide machines to track the desired path was designed. A Global Positioning System (GPS) was adopted in this algorithm to determine the current position of the machine. A path preview algorithm was designed to estimate the future position for the machine. On the basis of this system, dynamic tracking and display of the position was realised and the optimal path was provided through path planning.

Optimal stabilization and path-following controls for a bicycle

The article is about stabilizing and path-tracking control of a bicycle by a rider. It is based on previously published work, in which it has been shown how a driver's or rider's preview of the roadway can be combined with the linear dynamics of an appropriate vehicle to yield a problem of discrete-time optimal-linear-control-theory form. In the previous work, it was shown how an optimal ‘driver’ converts path preview sample values, modelled as deriving from a Gaussian white-noise process, into steering control inputs to cause the vehicle to follow the previewed path. The control compromises between precision and ease, to an extent that is controllable through choice of weights in the optimal control calculations. Research into the dynamics of bicycles has yielded a benchmark model, with equations of motion firmly established by extensive cross-checking. Model predictions have been verified for modest speeds by experimental testing. The established optimal linear preview stabilizing and tracking control theory is now brought together with the benchmark bicycle description to yield optimal controls for the bicycle for variations in speed and performance objectives. The resulting controls are installed in the bicycle, giving a virtual rider-controlled system, and frequency responses of the rider-controlled system are calculated to demonstrate tracking capability. Then path-tracking simulations are used to illustrate the behaviour of the controlled system. Tight and loose controls, representing different balances between tracking accuracy and control effort, are calculated and illustrated through the simulations.

Driver Steering Control and a New Perspective on Car Handling Qualities

The article is about steering control of cars by drivers, concentrating on following the lateral profile of the roadway, which is presumed visible ahead of the car. It builds on previously published work, in which it was shown how the driver's preview of the roadway can be combined with the linear dynamics of a simple car to yield a problem of discrete-time optimal-linear-control-theory form. In that work, it was shown how an optimal ‘driver’ of a linear car can convert the path preview sample values, modelled as deriving from a Gaussian white-noise process, into steering wheel displacement commands to cause the car to follow the previewed path with an attractive compromise between precision and ease. Recognizing that real roadway excitation is not so rich in high frequencies as white-noise, a low-pass filter is added to the system. The white-noise sample values are filtered before being seen by the driver. Numerical results are used to show that the optimal preview control is unaltered by the inclusion of the low-pass filter, whereas the feedback control is affected diminishingly as the preview increases. Then, using the established theoretical basis, new results are generated to show time-invariant optimal preview controls for cars and drivers with different layouts and priorities. Tight and loose controls, representing different balances between tracking accuracy and control effort, are calculated and illustrated through simulation. A new performance criterion with handling qualities implications is set up, involving the minimization of the preview distance required. The sensitivities of this distance to variations in the car design parameters are calculated. The influence of additional rear wheel steering is studied from the viewpoint of the preview distance required and the form of the optimal preview gain sequence. Path-following simulations are used to illustrate relatively high-authority and relatively low-authority control strategies, showing manoeuvring well in advance of a turn under appropriate circumstances. The results yield new insights into driver steering control behaviour and vehicle design optimization. The article concludes with a discussion of research in progress aimed at a further improved understanding of how drivers control their vehicles.

Adapting an Articulated Vehicle to its Drivers

A yaw plane model with limited roll-DOF of a five-axle tractor semitrailer is developed to study the open-loop directional dynamics of the vehicle. A driver model incorporating path preview, low and high frequency compensatory gains and time delays, and prediction of the vehicle state, is developed and integrated with the vehicle model. The coupled model is analyzed to investigate the vehicle design, which could be best adapted in view of the control limits of different driver, which are identified in terms of preview distance, reaction time and compensatory gain. A performance index based upon the vehicle path tracking, directional response characteristics and the driver’s steering effort is formulated and minimized using Gauss-Newton method to derive the desirable ranges of vehicle parameters, that could be adapted for drivers with varying skills. It is concluded that the adaptability and thus the directional performance of the vehicle can be enhanced through variations in the weights and dimensions, and compliant properties of the suspension, tire and the fifth wheel. The results of the study suggest that a driver with superior driving skill can easily adapt a vehicle with relatively large size, soft suspension and higher degree of oversteer. The results further show that the driver-adapted vehicle yields up to 33 percent reduction in the steering effort demand posed on the driver, while the roll angle and yaw rate response decrease by up to 40 percent.

Cutter path preview for contour machining based on efficient identification of motion error sources

This paper presents an efficient approach to the identification of the motion error sources in CNC contour machining. The motion error sources include the mismatch of feed drives, the friction disturbances and the cutter path curvature. Their effects on the contour error in linear and circular motions are analyzed, respectively. On the basis of the analysis, the procedure for the identification of these error-source effects is developed. With the identification results, the cutter path trajectory can be previewed before the real run.