Effective Parameters, Real Behavior: Renormalization for Robotics -- From Infinite Electron Mass to Sim-to-Real Gap
2026-07-27 • Robotics
Robotics
AI summaryⓘ
The authors suggest a new way to make robot simulations behave more like real life by adjusting some parameters to cover for what the simulator misses. Instead of making simulators super detailed, they use 'effective' parameters that change depending on how the simulation is run, kind of like tuning a recipe to get the right taste. They explain this with examples like robot control and swimming, showing how these parameter tweaks can mimic real-world physics. They also provide a method to pick what to measure and how to set these effective parameters to improve simulations.
sim-to-real gaprenormalizationeffective parametersPD controlsimulation frequencydynamic manipulationunderwater roboticsparameter tuningrobot simulation
Authors
Youran Sun, Jiaxuan Guo, Xingyu Ren, Chugang Yi, Haizhao Yang
Abstract
Bridging the sim-to-real gap is a central problem in robotics, and the prevailing approach is to build increasingly accurate simulators. Here, we propose another approach based on renormalization: using effective, resolution-dependent parameters to absorb details omitted by the simulator and reproduce real behavior. These parameters may differ from measured physical values because they compensate for what the simulator leaves out. We demonstrate this mechanism analytically for proportional--derivative (PD) control at finite simulation frequency, where proportional feedback changes the effective derivative gain and derivative feedback changes the effective inertia. We then interpret dynamic rope manipulation and underwater swimming through the same perspective. Finally, we present a practical procedure for choosing observables, identifying omitted physics, and determining effective parameters. Renormalization offers robotics a complementary path across the sim-to-real gap: effective parameters, real behavior.