Novel automatic parking system vehicle positioning algorithm

Abstract

The invention discloses a novel vehicle positioning algorithm for an automatic parking system. The novel vehicle positioning algorithm comprises the following steps of 1, an Ackerman steering model of Ackerman offset being considered in a dead reckoning algorithm; 2, the steering transmission ratio having the nonlinear characteristic, and fitting being conducted through a quadratic nonlinear function; 3, calculating course angle increment in a dead reckoning algorithm by adopting the number of front wheel pulses; 4, calculating the unit pulse mileage in the dead reckoning algorithm by adopting a straight line and an average value calibrated under the minimum turning radius; and 5, for error correction of dead reckoning, a visual or ultrasonic sensor being adopted to identify relative parking space information, and an error of the current vehicle pose being corrected. The algorithm model is simple, the novel dead reckoning algorithm is designed based on the mileage pulse number, the vehicle body parameters and the vehicle steering angle information and considering the Ackerman offset, large computing power is not occupied, engineering application is facilitated, meanwhile, the application range is wide, and cost is low.

Description

technical field [0001] The invention relates to the technical field of automatic driving, in particular to a novel vehicle positioning algorithm for an automatic parking system. Background technique [0002] At present, image perception technology and sensor technology are developing rapidly, and along with the development needs of automation and intelligence in the automobile industry, automatic parking systems have become a function that major automobile manufacturers are vying to equip. The automatic parking system mainly includes a positioning module, a parking space detection module, a decision-making module, a planning control module and a human-computer interaction module. After the user first enables the automatic parking function through human-computer interaction, the positioning module starts real-time dead reckoning to calculate the vehicle pose; the second step, the vehicle detects the parking space through image perception technology or ultrasonic radar technol...

AI Technical Summary

Problems solved by technology

This algorithm is simple, the algorithm itself is relatively ideal, and does not consider the steering performance of the vehicle. Generally, in order to ensure maneuverability and reduce tire wear, the steering of the vehicle is not completely Ackerman steering that meets the standard, that is, the Ackerman rate is not equal to 100. %, in addition, changes such as wheel slippage and vehicle rolling radius will also affect the accuracy of the model, as the mileage increases, the positioning error will gradually increase, and the positioning error is not corrected, resulting in poor positioning accuracy

Another solution is to obtain the yaw angle of the vehicle based on the inertial sensor, and correct the pose of the above-mentioned first method through the Kalman filter, thereby improving the positioning accuracy. This algorithm is more complicated and increases the computational burden of the controller, and the inertia Sensors add cost

Images