The invention discloses an automatic driving behavior decision-making method based on deep reinforcement learning, and relates to the technical field of unmanned driving, and the method comprises the following steps: S1, determining a real-time road condition; s11, determining the running state of the vehicle body; s2, special signal acquisition and discrimination: S21, special vehicle signal acquisition; s22, determining the position of a special vehicle; s3, adjusting the posture of the vehicle body: S31, determining the position of the vehicle body; s32, vehicle body avoiding space is determined; s33, a vehicle body avoiding path is selected; s4, vehicle driving state confirmation: S41, special vehicle position departure determination; s42, confirming a driving path of the vehicle body; and S43: data processing, data storage, data uploading and deep learning. According to the method, judgment and deep learning are carried out on a special vehicle condition processing method during unmanned driving, the timeliness and accuracy of automatic driving behavior decision making under unmanned driving are optimized, and the adaptive capacity of an unmanned vehicle in a complex traffic environment is improved.
本发明公开了一种基于深度强化学习的自动驾驶行为决策方法,涉及无人驾驶技术领域,包括以下步骤:S1:实时路况确定;还包括S11:车身行驶状态确定;S2:特殊信号采集判别:包括S21:特殊车辆信号采集;S22:特殊车辆位置确定;S3:车身姿态调整:包括S31:车身位置确定;S32:车身避让空间确定;S33:车身避让路径选择;S4:车辆行驶状态确认:包括S41:特殊车辆位置远离确定;S42:车身行驶路径确认;S43:数据处理、数据存储、数据上传和深度学习。本发明解决了对无人驾驶时遇到特种车辆情况处理办法进行判定和深度学习,优化了无人驾驶下自动驾驶行为决策的及时性和准确性,提高了无人驾驶车辆在复杂交通环境下的适应能力。
Automatic driving behavior decision-making method based on deep reinforcement learning
基于深度强化学习的自动驾驶行为决策方法
14.05.2024
Patent
Elektronische Ressource
Chinesisch
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion |
Automatic driving behavior decision-making method based on deep reinforcement learning
Europäisches Patentamt | 2020
|Automatic driving decision-making method based on deep reinforcement learning
Europäisches Patentamt | 2024
|Automatic driving decision-making system based on deep reinforcement learning
Europäisches Patentamt | 2024
|Automatic driving behavior decision-making method and system based on reinforcement learning
Europäisches Patentamt | 2024
|Automatic driving lane keeping decision-making method based on deep reinforcement learning
Europäisches Patentamt | 2025
|