arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.04590cs.ROcs.AIcs.LGcs.SYeess.SY
收藏

深度模型预测优化

Deep Model Predictive Optimization

University of Washington(华盛顿大学);Robotics Institute, Carnegie Mellon University(卡内基梅隆大学机器人研究所)

Jacob Sacks, Rwik Rana, Kevin Huang, Alex Spitzer, Guanya Shi, Byron Boots

AI总结:

本文提出深度模型预测优化(DMPO),通过经验直接学习MPC优化算法的内循环,在四旋翼飞行器轨迹跟踪任务中以更少样本和内存超越基线MPC与端到端强化学习策略,并具备零样本抗风扰适应能力。

摘要:

机器人领域的一个主要挑战是设计能够在现实世界中实现复杂且敏捷行为的鲁棒策略。在一个极端,我们有无模型强化学习(MFRL),它极其灵活且通用,但往往会产生脆弱的策略。相比之下,模型预测控制(MPC)在每个时间步持续重新规划,以保持对扰动和模型不准确性的鲁棒性。然而,尽管MPC在现实世界中取得了成功,但它往往表现不如最优策略。这是由于模型质量、短规划视野导致的短视行为以及计算约束带来的近似。而且即使拥有完美的模型和足够的计算资源,MPC也可能陷入较差的局部最优,这在很大程度上取决于优化算法的质量。为此,我们提出了深度模型预测优化(DMPO),它通过经验直接学习MPC优化算法的内循环,专门针对控制问题的需求进行定制。我们在真实四旋翼飞行器的敏捷轨迹跟踪任务上评估了DMPO,在给定的计算预算下,它比基线MPC算法提高了性能。它可以用更少的样本超越最佳MPC算法最多27%,并且比使用MFRL训练的端到端策略高出19%。此外,由于DMPO需要的样本更少,它还可以用少4.3倍的内存实现这些优势。当我们让四旋翼飞行器在带有附加阻力板的湍流风场中飞行时,DMPO可以零样本适应,同时仍然优于所有基线。更多结果可在 https://tinyurl.com/mr2ywmnw 找到。

英文摘要:

A major challenge in robotics is to design robust policies which enable complex and agile behaviors in the real world. On one end of the spectrum, we have model-free reinforcement learning (MFRL), which is incredibly flexible and general but often results in brittle policies. In contrast, model predictive control (MPC) continually re-plans at each time step to remain robust to perturbations and model inaccuracies. However, despite its real-world successes, MPC often under-performs the optimal strategy. This is due to model quality, myopic behavior from short planning horizons, and approximations due to computational constraints. And even with a perfect model and enough compute, MPC can get stuck in bad local optima, depending heavily on the quality of the optimization algorithm. To this end, we propose Deep Model Predictive Optimization (DMPO), which learns the inner-loop of an MPC optimization algorithm directly via experience, specifically tailored to the needs of the control problem. We evaluate DMPO on a real quadrotor agile trajectory tracking task, on which it improves performance over a baseline MPC algorithm for a given computational budget. It can outperform the best MPC algorithm by up to 27% with fewer samples and an end-to-end policy trained with MFRL by 19%. Moreover, because DMPO requires fewer samples, it can also achieve these benefits with 4.3X less memory. When we subject the quadrotor to turbulent wind fields with an attached drag plate, DMPO can adapt zero-shot while still outperforming all baselines. Additional results can be found at https://tinyurl.com/mr2ywmnw.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑