Research On Modelling And Simulation Of The Parking Car Axle Slant Angle Based On The Tire Outline Identification

With the fast developing of social economy, more and more motor vehicles are possessed. Moreover, the high speed development of the automotive technology makes the running speed faster and faster. The frequencies of middle, high speed motor vehicles crossing are becoming higher and higher, such that the improper headlamp beam direction has been one of the mains factors which influence the safety running of vehicles at night. However the current headlamp detection in our country is hard to satisfy the security demanding of the increasing automotive industry. In the headlamp detection the position of headlamp is supposed to be assigned with the stopped car body. But in real application, with out correction device in the detection system, it is easy to misjudge the car body location only by the towing guider's objective feeling. There are two factors causing the detection error: 1) the requirement of the headlamp beam direction being perpendicular to the horizontal rails of the detector; 2) The distance between the headlamp and the detector. The first factor is the main topic of this thesis, which is to overcome the error caused by angle inclination between the headlamp beam direction and the horizontal rails of the detector.At present, the detection baseline is the car body axis, which is corrected by adjusting device to be perpendicular to the horizontal rails of the detector. But when the rear axle is not perpendicular to the body axis, the baseline is not the running direction any more or the headlamp beam direction, which resulted in the detection inaccuracy. Therefore this thesis focuses on taking the running direction as the detection baseline, and aims at requiring the slant angle of the running direction, in order to updating the detection result.Because the running direction is decided by the rear axle, and the slant angle of trailing wheel is the slant angle of the running direction, in this thesis, the slant angle is analyzed and estimated by obtaining the trailing wheel outline.The main contents are presented in the thesis:1. The sidewall data are utilized to fit a straight line, and it is verified that the slope of the line is the tangent value of the slant angle of the running direction. Mathematical proof is given under ideal condition. In practical application the sidewall outline is not axial symmetry. It will cause the discrepancy between the slope of the fitted line and the tangent value, which needs further error analysis of numerical simulation and experiment.2. The hard ware design of physical construction for error correction device is designed based on Solidworks, such that the wheel outline can be obtained automatically. On the horizontal axis and vertical axis are the data obtained by the wire draw sensor and by laser range finder respectively. Hence when car stops on the pointed stop line, the trailing wheel outline can be obtained from computer by letting the device move and scan the whole wheel.3. Simulate analysis is applied by Matlab. First the shape and size of the tire is summarized and analyzed, and the wheel model is built using Matlab via setting governing equation, mirror image, rotation etc method. The model equation is simplified by setting the tire center be the origin of coordinate, and letting the tire turn angle be zero.4. The estimated points on the tire outline are denser and the quantity is much more than that of the measured data. Coordinate conversation is performed convert the curve slang angle into from 0. At last the simulated measured data are obtained via mirror image, interpolation etc processing methods.5. Remove the nonutility data and kept valuable data which are from the sidewall. The cut-off points of the two sidewall ends are obtained by using the characteristics like numerical mutation, gradient change, and optimizing the center of circle, radius etc. The data between the cut-off points are the required data.6. The data of sidewall are extracted, and are fitted into a straight line, of which the slope is the tangent value of the slant angle.7. Different cases are analyzed. The factors like The tangent value is estimated by changing the shape and size of tire, the hole number between spokes, the size and position of convex body, amount, height of range finder, the inclination of car. Simultaneously the error is analyzed in order to analyze the influence of each factor to the slant angle identification.8. Experiment has been done on different height with identical angle and identical angle with different angles respectively. The measured data are polluted by the hand operation, vibration and noise. Although the relation between points are not in a regular order, the whole trend direction is reasonable and in a certain order for the outline. Therefore, multiple points are used to fit the line and calculate the derivative at each point. Then base on several groups experiment, threshold value is decided by statistic method. Such that the cut-off points can be got and the data of sidewall are extract to obtain the slant value. Currently the main experiment cases are performed corresponding to the slant angle value of the running direction 0°,1°,2°,3°,4°,5°. The errors of the angle identification in the 6 cases are compared and analyzed. The experiment error analysi verifies that the proposed axle slant angle identification based on the tire outline is feasible.