arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.26341cs.RO
收藏

扩展非线性优化:多问题单GPU

Scaling Nonlinear Optimization: Many Problems One GPU

Department of Mechanical Engineering, University of California, Berkeley(加州大学伯克利分校机械工程系);Department of Biosystems Science and Engineering, ETH Zürich(苏黎世联邦理工学院生物系统科学与工程系)

John Viljoen, Johanna Haffner, Lasse Peters, Masayoshi Tomizuka, Negar Mehr

AI总结:

提出首个基于JAX的GPU批处理NLP求解器jaxipm,通过异构迭代融合和迭代级批处理实现多问题并发求解,在四旋翼控制任务中吞吐量提升达32.85倍。

摘要:

许多机器人问题,包括轨迹优化、逆运动学和接触丰富的运动规划,都归结为非线性规划(NLP)。成熟的NLP求解器如IPOPT可以解决这些问题,提供硬约束满足、最优性保证以及随问题维度良好的扩展性。这些求解器支撑着机器人学中的梯度方法,但仍然是CPU绑定的,并且一次只求解一个问题,阻碍了它们集成到GPU批处理学习流程中。另一方面,基于采样的方法,如强化学习、模型预测路径积分和模仿学习,已成为现代机器人学研究的核心,因为它们能够利用GPU批处理模拟器。这些模拟器每秒可以生成比以往多几个数量级的动力学滚动。如果存在一个GPU批处理NLP求解器,它将解锁在每秒生成的约束局部最优解数量上的类似加速。这种同时求解多个问题与一次求解单个问题的模式,是将NLP求解器集成到现代GPU批处理机器人框架中的关键要求。为此,我们介绍了\texttt{jaxipm},第一个基于IPOPT并在JAX中实现的GPU批处理NLP求解器。我们通过重新设计IPOPT的算法,使用\textit{异构迭代融合}消除控制流,并通过\textit{迭代级批处理}最小化GPU空闲时间来实现这一点。我们在各种四旋翼非线性模型预测控制基准上评估了\texttt{jaxipm},包括存在障碍物时的参考跟踪、多四旋翼无碰撞导航以及杂乱环境中的导航。我们展示了相比IPOPT高达$32.85\times$的吞吐量提升。我们的完整开源代码库可在此https URL获取。

英文摘要:

Modern GPUs and software frameworks have made sampling-based methods, such as reinforcement learning and model predictive path integral control, increasingly popular in robotics. However, the logic these methods can execute per sample is limited to what has been implemented on the GPU. Nonlinear programs (NLPs) are one of the most expressive tools in robotics, reliably minimizing a cost while exactly enforcing general constraints. NLP optimizers were among the field's main workhorses, but were not designed for modern GPU hardware. Embedding NLPs such as trajectory optimization and inverse kinematics directly within each sample would be highly valuable, yet the only current option is to call a CPU solver per sample, possibly in parallel on a multi-core CPU. This bottlenecks the CPU, requires GPU-to-CPU-to-GPU transfers when the rest of the pipeline resides on the GPU, and leaves the GPU intermittently idle. To address this issue, in this work we present jaxipm, the first GPU-batched solver for generally constrained nonlinear optimization problems. Our approach is a JAX implementation of the battle-tested IPOPT algorithm, facilitated by eliminating control flow through heterogeneous iteration fusion, and minimizing GPU idle time through iteration-level batching. Our implementation enables solving hundreds of NLPs in parallel on a single GPU. We validate jaxipm on a range of quadrotor nonlinear model predictive control benchmarks, where on a single GPU we achieve between 30.7% and 69.9% of the throughput of 64 IPOPT instances running in parallel on a 64-core CPU. open-source codebase: https://github.com/johnviljoen/jaxipm

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。

↑