当前位置: X-MOL 学术ACM Trans. Model. Comput. Simul. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Transfer Reinforcement Learning for Autonomous Driving
ACM Transactions on Modeling and Computer Simulation ( IF 0.9 ) Pub Date : 2021-07-18 , DOI: 10.1145/3449356
Aravind Balakrishnan 1 , Jaeyoung Lee 1 , Ashish Gaurav 1 , Krzysztof Czarnecki 1 , Sean Sedwards 1
Affiliation  

Reinforcement learning (RL) is an attractive way to implement high-level decision-making policies for autonomous driving, but learning directly from a real vehicle or a high-fidelity simulator is variously infeasible. We therefore consider the problem of transfer reinforcement learning and study how a policy learned in a simple environment using WiseMove can be transferred to our high-fidelity simulator, W ise M ove . WiseMove is a framework to study safety and other aspects of RL for autonomous driving. W ise M ove accurately reproduces the dynamics and software stack of our real vehicle. We find that the accurately modelled perception errors in W ise M ove contribute the most to the transfer problem. These errors, when even naively modelled in WiseMove , provide an RL policy that performs better in W ise M ove than a hand-crafted rule-based policy. Applying domain randomization to the environment in WiseMove yields an even better policy. The final RL policy reduces the failures due to perception errors from 10% to 2.75%. We also observe that the RL policy has significantly less reliance on velocity compared to the rule-based policy, having learned that its measurement is unreliable.

中文翻译:

自动驾驶的迁移强化学习

强化学习 (RL) 是实施自动驾驶高级决策策略的一种有吸引力的方式,但直接从真实车辆或高保真模拟器中学习是不可行的。因此,我们考虑迁移强化学习的问题,并研究策略如何在简单的环境中使用智动可以转移到我们的高保真模拟器 W伊势.智动是研究自动驾驶强化学习的安全性和其他方面的框架。W伊势准确地再现了我们真实车辆的动力学和软件堆栈。我们发现 W 中准确建模的感知误差伊势对转移问题的贡献最大。这些错误,即使是天真地建模智动,提供在 W 中表现更好的 RL 策略伊势而不是手工制定的基于规则的策略。将域随机化应用于环境智动产生更好的政策。最终的 RL 策略将由于感知错误导致的失败率从 10% 减少到 2.75%。我们还观察到,与基于规则的策略相比,RL 策略对速度的依赖要少得多,因为我们了解到它的测量是不可靠的。
更新日期:2021-07-18
down
wechat
bug