首页|考虑舒适度的智能汽车人工蜂群轨迹规划方法

考虑舒适度的智能汽车人工蜂群轨迹规划方法

扫码查看
原文链接
  • 万方数据
  • 维普
为提高智能汽车在避障换道过程中的舒适度与安全性，提出了一种考虑舒适度的智能汽车人工蜂群轨迹规划方法。通过采样法在速度-时间空间内进行位置采样，得到基于时间的速度序列，并设计新的蜜源搜索策略和蜜源更新策略以提高人工蜂群算法的搜索精度和收敛速度。结合五次多项式拟合速度序列与蜜源位置，得到换道轨迹。考虑车辆加速度和冲击度对舒适度的影响，设计轨迹舒适度评价函数并融合到适应度函数中，通过碰撞检测结果优化换道轨迹。基于Simulink-PreScan-CarSim联合仿真平台验证算法有效性。结果表明:面对不同工况，本文提出方法能够规划出符合加速度、冲击度最值约束的无碰撞安全换道轨迹，且规划结果优于未考虑舒适度的规划轨迹。在实车测试中，本文所提方法在低速场景下能够实现对静态障碍物、低速动态障碍物的避障轨迹规划，避障轨迹均符合舒适度要求，且表现出较好的可跟踪性。
Artificial bee colony trajectory planning algorithm for intelligent vehicles considering comfortable
To enhance the comfort and safety of intelligent vehicles during obstacle avoidance and lane changing,a trajectory planning method was proposed using an artificial bee colony algorithm that considers comfort.By sampling positions in the speed-time space,a time-based speed sequence was obtained.New honey source search and update strategies were designed to improve the search accuracy and convergence speed of the artificial bee colony algorithm.The lane-changing trajectory was derived by combining a fifth-order polynomial fit of the speed sequence with honey source positions.Taking into account vehicle acceleration and jerk for comfort,a trajectory comfort evaluation function was designed and integrated into the fitness function to optimize the lane-changing trajectory based on collision detection results.The effectiveness of the algorithm was validated using the Simulink-PreScan-CarSim jointsimulation platform.The results show that under various conditions,the proposed method can plan collision-free lane-changing trajectories that comply with acceleration and jerk constraints,outperforming trajectories planned without considering comfort.In real vehicle tests,the proposed method can achieve obstacle avoidance trajectory planning for static obstacles and low-speed moving obstacles in low-speed scenarios,with all trajectories meeting comfort requirements and demonstrating good trackability.

vehicle engineeringintelligent vehicletrajectory planningcomfort evaluationartificial bee colonyvelocity planningsafety collision detection

谢宪毅、张明君、金立生、周彬、胡涛、白宇飞

展开 >

燕山大学 车辆与能源学院,河北 秦皇岛 066004

清华大学 汽车安全与节能国家重点实验室,北京 100084

中国汽车技术研究中心有限公司 中汽研汽车检验中心(天津)有限公司,天津 300300

北京航空航天大学 车路一体智能交通全国重点实验室,北京 102206

展开 >

车辆工程 智能汽车 轨迹规划 舒适度评价 人工蜂群 速度规划 安全碰撞检测

国家自然科学基金项目清华大学汽车安全与节能国家重点实验室开放基金项目河北省省级科技计划项目河北省省级科技计划项目国家重点研发计划项目

52072333KFY2211F2021203107F20222030542022YFF0604901

2024

10.13229/j.cnki.jdxbgxb.20231280

吉林大学学报(工学版)
吉林大学

吉林大学学报(工学版)

CSTPCD北大核心
影响因子：0.792
ISSN：1671-5497
年,卷(期)：2024.54(6)
  • 7