汽车工程

• •    下一篇

基于驾驶员经验的自动泊车规控算法研究

胡文1, 谭运生2, 康龙云1, 于宗光3   

  1. (1.华南理工大学电力学院,广州,510640;2.领途汽车有限公司智能驾驶开发部,无锡,214000; 3.中国电子科技集团公司第五十八研究所,无锡,214000
  • 出版日期:2019-03-05 发布日期:2019-03-05
  • 基金资助:
     

Automatic Parking Planning Algorithm Based on Driver’s Parking Experience

    

  1.  
  • Online:2019-03-05 Published:2019-03-05
  • Supported by:
     

摘要: 为解决现有泊车算法中的原地转向和泊车实时性问题,以及还存在一定的实时性和精确度不足问题,本文在借鉴熟练驾驶员泊车经验的基础上,对各泊车过程进行机理分析,并建立各相应的数学模型,设计了一种新的自动泊车规划算法。该算法主要有泊车控制、转向控制和速度控制三个部分:泊车控制用于规划泊车路径及输出转向角和期望速度;转向控制用于控制方向盘转向时机;速度控制用于控制车辆在泊车过程中的速度要求。为满足自动泊车系统的实时性和精确性要求,控制算法在设计过程中,采用车辆运动学分析和几何规划的方法对控制器进行计算设计。仿真结果表明:文本所提算法能够实现精确的自动泊车过程,其既能满足精度要求,又能满足泊车控制的实时性要求。

关键词: 自动泊车, 路径规划, 驾驶经验, 平行泊车

Abstract: In order to solve pivot steering issue, real-time and accuracy issue of automatic parking system, this paper use the parking experience of skilled drivers and carries out the parking process Mechanism analysis, and establish corresponding mathematical models, designed a new automatic parking planning algorithm. The algorithm mainly has three parts: parking motion plan, steering control and speed control; parking control is used to plan the parking path and output steering angle and desired speed; steering control is used to control the steering wheel steering timing; speed control is used to control the vehicle in Speed requirements during parking. In order to meet the real-time and accuracy requirements of the automatic parking system, in the design process, the parking trajectory is planned and designed based on the excellent driver parking experience and geometric planning method, and the vehicle kinematics analysis is used for stability control design of steering and driving speed. The design results are simulated, and other algorithms are compared. The results show that the proposed algorithm has good real-time and accuracy.

Key words: automatic parking, path planning, driving experience, parallel parking

中图分类号: 

  •