人工智能培训

搜索

机器学习论文:基于障碍物危险的强化学习行人路径规划模型(A pedestrian path-planning model in accordance with

[复制链接]
zhq8008 发表于 2019-12-9 13:16:48 | 显示全部楼层 |阅读模式
zhq8008 2019-12-9 13:16:48 224 0 显示全部楼层
机器学习论文:基于障碍物危险的强化学习行人路径规划模型(A pedestrian path-planning model in accordance with obstacle's danger  with reinforcement learning)大多数微观行人导航模型使用应用于行人代理的“力”的概念来复制导航环境。尽管该方法在常规情况下可以提供令人信服的结果,但在许多典型环境中并不总是类似于自然的行人导航行为。在我们的研究中,我们提出了一种使用强化学习的新方法来模拟行人代理路径规划和避碰问题。这种方法的主要重点是利用人类对环境的感知和对干扰的危险意识。我们模型的实现表明,代理人计划的路径在某些方面与人类行人有很多相似之处,例如遵循常见的步行惯例和人类行为。
Most microscopic pedestrian navigation models use the concept of "forces"applied to the pedestrian agents to replicate the navigation environment.Whilethe approach could provide believable results in regular situations, it doesnot always resemble natural pedestrian navigation behaviour in many typicalsettings.In our research, we proposed a novel approach using reinforcementlearning for simulation of pedestrian agent path planning and collisionavoidance problem.The primary focus of this approach is using human perceptionof the environment and danger awareness of interferences.The implementation ofour model has shown that the path planned by the agent shares many similaritieswith a human pedestrian in several aspects such as following common walkingconventions and human behaviours.机器学习论文:基于障碍物危险的强化学习行人路径规划模型(A pedestrian path-planning model in accordance with obstacle's danger  with reinforcement learning)
URL地址:https://arxiv.org/abs/1912.02945     ----pdf下载地址:https://arxiv.org/pdf/1912.02945    ----机器学习论文:基于障碍物危险的强化学习行人路径规划模型(A pedestrian path-planning model in accordance with obstacle's danger  with reinforcement learning)
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则 返回列表 发新帖

zhq8008当前离线
新手上路

查看:224 | 回复:0

快速回复 返回顶部 返回列表