Robotics Research Hub
Draft: A Parametric Tool for Robot Design Exploration

Draft: A Parametric Tool for Robot Design Exploration

David Nguyen, Marcelo Coelho, Sangbae Kim · N/A · 2026

locomotion reinforcement-learning control
PDF arXiv

Framework

N/A

License

N/A

Stars

N/A

Summary

Robot performance is often limited by the cost of iterating on morphology and control together, since every computer-aided design (CAD) change has to be carried into a simulation-ready model before control work begins.

Abstract Summary

Robot performance is often limited by the cost of iterating on morphology and control together, since every computer-aided design (CAD) change has to be carried into a simulation-ready model before control work begins. Co-design methods attempt to close this gap, but each uses a model generator written for a single platform or lack the use of real-world data to suggest that designs are plausible. We present Draft, a parametric generation tool whose generalized engine compiles any parametric tree of serial chains into a simulation-ready MJCF model, without CAD. It allows engineers to explore design tradeoffs through easily adjustable models and evaluate how changes influence controller performance. Draft grounds the free parameters of each design using trends fitted to a survey of $114$ actuators and $49$ published robot descriptions, so that a generated robot is anchored to real-world hardware. We validate those trends wholistically by building twins of four off-the-shelf robots, whose masses agree to $1.10\times$ geometric mean fold error. Finally, we demonstrate how Draft exposes design tradeoffs by evaluating three quadrupeds through a two-stage reinforcement learning curriculum.

Key Points

  • Robot performance is often limited by the cost of iterating on morphology and control together, since every computer-aided design (CAD) change has to be carried into a simulation-ready model before control work begins.
  • Co-design methods attempt to close this gap, but each uses a model generator written for a single platform or lack the use of real-world data to suggest that designs are plausible.
  • We present Draft, a parametric generation tool whose generalized engine compiles any parametric tree of serial chains into a simulation-ready MJCF model, without CAD.
  • It allows engineers to explore design tradeoffs through easily adjustable models and evaluate how changes influence controller performance.
  • Draft grounds the free parameters of each design using trends fitted to a survey of $114$ actuators and $49$ published robot descriptions, so that a generated robot is anchored to real-world hardware.

Parametric Model Generation

In our formulation, a robot is defined by three files. The first is a parameter file that specifies values for link lengths, densities, and motor classes etc. The second is a kinematic tree whose fields are arithmetic expressions which correspond to the parametric file’s values. The third describes the root body mesh.

Because the tree holds expressions rather than numbers, it can be written so that fewer parameters have to be specified. A limb’s radius can be linked to the motor that drives it rather than manually specified.

Parametric Model Generation — Draft: A Parametric Tool for Robot Design Exploration
Fig. 2: All generated Draft models are constructed from four main link primitives. The basic link includes a link and motor at its distal end, the sphere link terminates articulated chains with a contact point, the foot link does the same but with a support surface, and the root link holds custom geometry for limbs to extend from.

Fitting Model Parameters to Hardware

The generator guarantees that a model is internally consistent, not that the design is plausible. Nothing in Sec. III stops a sweep from asking for an actuator with twice the torque density of anything on the market, and would report performance on impossible hardware. We therefore fit equations to surveys of real hardware and compile them back into the generation pipeline as guard rails.

We perform two surveys, one on actuators and another on robot structure. Every actuator fit is scored the same way, using leave-one-out (LOO) validation.

Fitting Model Parameters to Hardware — Draft: A Parametric Tool for Robot Design Exploration
Fig. 3: An actuator’s parameters are derived from data-informed relations. The inputs are peak torque $\tau$, no-load speed $\omega_{\text{NL}}$, and aspect ratio $a$. Mass $m$ is estimated from $\tau$, and gear ratio $N$ from $\omega_{\text{NL}}$. Volume $V$ is estimated from $\tau$ and $N$; together with $a$, it determines radius $r$ and length $\ell$. Finally, the rotor inertia $J$ is derived from $r$.

Evaluating Designs with Reinforcement Learning

Our model generation scheme is only useful if it shows how varying parameters influences controller performance. We demonstrate this, by generating quadrupeds with three distinct parameter sets ($\boldsymbol{\theta}_{c},\boldsymbol{\theta}_{g},\boldsymbol{\theta}_{b}$) and training them under one curriculum.

To validate the robustness of the policies, we also perturb the inertial parameters post-training and analyze the results. Each quadruped design in Fig.

Evaluating Designs with Reinforcement Learning — Draft: A Parametric Tool for Robot Design Exploration
Fig. 4: The four relations that dictate gear ratio, mass, volume, and rotor inertia plotted against the raw data. Panel (a) shows the power-law fit for gear ratio against no-load actuator speed, while (b) shows the fit for actuator mass against peak output torque. Panels (c) and (d) show prediction performance for volume and rotor inertia, respectively. Points are grouped by actuator class (QDD, MidGear, and High GR) and show agreement with the fitted predictions (black lines). The blue bands are $\pm 25\%$ in (c) and (d).

Sources and Demonstrations

Figures are reproduced from Nguyen et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.

Share

Related Papers

Adaptive Rejection of Unknown Hybrid Sinusoidal Disturbances with Application to Humanoid Walking under Unknown Ground Motion
arXiv preprint

Adaptive Rejection of Unknown Hybrid Sinusoidal Disturbances with Application to Humanoid Walking under Unknown Ground Motion

I-Chia Chang, Jacob Stewart, Petros A. Ioannou et al.

Legged robot control is not yet practical in non-inertial environments, where the ground and surroundings accelerate in the inertial Earth frame, mainly because of the unknown, time-varying, persistent perturbations caused by environmental motion, compounded by the associated hybrid robotic dynamics.

locomotion reinforcement-learning control
PDF Intermediate