# Accelerated co-design of robots through morphological pretraining

arXiv: https://arxiv.org/abs/2502.10862 · HTML: https://mlsanity.com/paper/2502.10862/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 15 Feb 2025.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 12 of 12)

- [Embodied Intelligence via Learning and Evolution](https://mlsanity.com/paper/2102.02202/) (2021)
- [One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Control](https://mlsanity.com/paper/2007.04976/) (2020)
- [MetaMorph: Learning Universal Controllers with Transformers](https://mlsanity.com/paper/2203.11931/) (2022)
- [One Policy to Run Them All: an End-to-end Learning Approach to Multi-Embodiment Locomotion](https://mlsanity.com/paper/2409.06366/) (2024)
- [Scalable Co-Optimization of Morphology and Control in Embodied Machines](https://mlsanity.com/paper/1706.06133/) (2017)
- [Transform2Act: Learning a Transform-and-Control Policy for Efficient Agent Design](https://mlsanity.com/paper/2110.03659/) (2021)
- [Sim-to-Real of Soft Robots with Learned Residual Physics](https://mlsanity.com/paper/2402.01086/) (2024)
- [Efficient automatic design of robots](https://mlsanity.com/paper/2306.03263/) (2023)
- [Generating Freeform Endoskeletal Robots](https://mlsanity.com/paper/2412.01036/) (2024)
- [4D topology optimization: Integrated optimization of the structure and self-actuation of soft bodies for dynamic motions](https://mlsanity.com/paper/2302.00905/) (2023)
- [Evolution and learning in differentiable robots](https://mlsanity.com/paper/2405.14712/) (2024)
- [A non-cubic space-filling modular robot](https://mlsanity.com/paper/2403.01323/) (2024)

## Built on by (most cited first, 6 of 6)

- [House of Dextra: Cross-embodied Co-design for Dexterous Hands](https://mlsanity.com/paper/2512.03743/) (2025)
- [Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challenges](https://mlsanity.com/paper/2512.04770/) (2025)
- [Shaping the Evolutionary Dynamics of Robot Morphology via Adaptive Control Learning](https://mlsanity.com/paper/2608.23100/) (2026)
- [Bridging Body and Brain: Gene-Driven Morphology--Control Co-Design](https://mlsanity.com/paper/2609.31329/) (2026)
- [Evolving Dexterous Robots from Scratch](https://mlsanity.com/paper/2609.33101/) (2026)
- [Physics-based Sphere Packing for Lagrangian Mesh Morphing](https://mlsanity.com/paper/2610.08880/) (2026)

## Abstract (arXiv)

The co-design of robot morphology and neural control typically requires using reinforcement learning to approximate a unique control policy gradient for each body plan, demanding massive amounts of training data to measure the performance of each design. Here we show that a universal, morphology-agnostic controller can be rapidly and directly obtained by gradient-based optimization through differentiable simulation. This process of morphological pretraining allows the designer to explore non-differentiable changes to a robot's physical layout (e.g. adding, removing and recombining discrete body parts) and immediately determine which revisions are beneficial and which are deleterious using the pretrained model. We term this process "zero-shot evolution" and compare it with the simultaneous co-optimization of a universal controller alongside an evolving design population. We find the latter results in diversity collapse, a previously unknown pathology whereby the population -- and thus the controller's training data -- converges to similar designs that are easier to steer with a shared universal controller. We show that zero-shot evolution with a pretrained controller quickly yields a diversity of highly performant designs, and by fine-tuning the pretrained controller on the current population throughout evolution, diversity is not only preserved but significantly increased as superior performance is achieved.