Amplify library enables reproducible trajectory optimization in robotics
Amplify: A Lightweight Library for Reproducible Nonlinear Programming Problems in Robotics
Robotics
Summary
Optimization helps robots figure out the best way to move or perform actions, but it’s often hard to repeat the exact steps and get the same answers again. The authors present Amplify, a small software library that makes it easier to describe and solve these robot movement problems in a way that others can reproduce. It uses a clear language to describe robot paths and physics directly inside the problem setup, making comparisons and reuse simpler. They tested Amplify against other similar software on robot walking and grasping tasks and shared how its design could help other tools become lighter and more repeatable.
What this means in practice
- •For robotics engineers: Solve robot motion planning problems with a library that embeds the dynamics and paths directly for reliable, repeatable optimization results.
- •For control system developers: Develop and test new control algorithms using a lightweight tool that transparently encodes optimization constraints making experiments easier to replicate.
Authors
Nelson Rosa
Abstract
Optimization problems (OPs) are key to solving many challenging research problems in robotics. However, reproducibility still remains a major issue. In this paper, we present Amplify, a lightweight nonlinear programming library aimed at reproducible results of robotic-related trajectory optimization problems. The minimalistic requirements for the 537-line library (80 characters per line) are an Internet connection, familiarity with the AMPL modeling language, and a text editor. Our primary contribution is the formulation of a library where trajectory optimization algorithms are represented directly within the optimization model. Specifically, we implement the algorithms used to compute the dynamics, trajectories, and reference motions as constraints of the OP in a declarative programming paradigm. We outline how our formulation of objectives, decisions variables, and constraints can be implemented in other transcription libraries that want to be lightweight and reproducible. We also compare the Amplify framework with 3 other libraries across examples of benchmark optimization problems across several fields, including bipedal locomotion and grasp planning.