SafeBimanual: Diffusion-based trajectory optimization for safe bimanual manipulation

Haoyuan Deng, Wenkai Guo, Qianzhun Wang, Zhenyu Wu, Ziwei Wang
Proceedings of The 9th Conference on Robot Learning, PMLR 305:3218-3238, 2025.

Abstract

Bimanual manipulation has been widely applied in household services and manufacturing, which enables the complex task completion with coordination requirements. Recent diffusion-based policy learning approaches have achieved promising performance in modeling action distributions for bimanual manipulation. However, they ignored the physical safety constraints of bimanual manipulation, which leads to the dangerous behaviors with damage to robots and objects. To this end, we propose a test-time trajectory optimization framework named SafeBimanual for any pre-trained diffusion-based bimanual manipulation policies, which imposes the safety constraints on bimanual actions to avoid dangerous robot behaviors with improved success rate. Specifically, we design diverse cost functions for safety constraints in different dual-arm cooperation patterns including avoidance of tearing objects and collision between arms and objects, which optimizes the manipulator trajectories with guided sampling of diffusion denoising process. Moreover, we employ a vision-language model (VLM) to schedule the cost functions by specifying keypoints and corresponding pairwise relationship, so that the optimal safety constraint is dynamically generated in the entire bimanual manipulation process. SafeBimanual demonstrates superiority on 8 simulated tasks in RoboTwin with a 11.1% increase in success rate and a 18.9% reduction in unsafe interactions over state-of-the-art diffusion-based methods. Extensive experiments on 4 real-world tasks further verify its practical value by improving the success rate by 32.5%.

Cite this Paper

BibTeX
@InProceedings{pmlr-v305-deng25c, title = {SafeBimanual: Diffusion-based trajectory optimization for safe bimanual manipulation}, author = {Deng, Haoyuan and Guo, Wenkai and Wang, Qianzhun and Wu, Zhenyu and Wang, Ziwei}, booktitle = {Proceedings of The 9th Conference on Robot Learning}, pages = {3218--3238}, year = {2025}, editor = {Lim, Joseph and Song, Shuran and Park, Hae-Won}, volume = {305}, series = {Proceedings of Machine Learning Research}, month = {27--30 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v305/main/assets/deng25c/deng25c.pdf}, url = {https://proceedings.mlr.press/v305/deng25c.html}, abstract = {Bimanual manipulation has been widely applied in household services and manufacturing, which enables the complex task completion with coordination requirements. Recent diffusion-based policy learning approaches have achieved promising performance in modeling action distributions for bimanual manipulation. However, they ignored the physical safety constraints of bimanual manipulation, which leads to the dangerous behaviors with damage to robots and objects. To this end, we propose a test-time trajectory optimization framework named SafeBimanual for any pre-trained diffusion-based bimanual manipulation policies, which imposes the safety constraints on bimanual actions to avoid dangerous robot behaviors with improved success rate. Specifically, we design diverse cost functions for safety constraints in different dual-arm cooperation patterns including avoidance of tearing objects and collision between arms and objects, which optimizes the manipulator trajectories with guided sampling of diffusion denoising process. Moreover, we employ a vision-language model (VLM) to schedule the cost functions by specifying keypoints and corresponding pairwise relationship, so that the optimal safety constraint is dynamically generated in the entire bimanual manipulation process. SafeBimanual demonstrates superiority on 8 simulated tasks in RoboTwin with a 11.1% increase in success rate and a 18.9% reduction in unsafe interactions over state-of-the-art diffusion-based methods. Extensive experiments on 4 real-world tasks further verify its practical value by improving the success rate by 32.5%.} }
Endnote
%0 Conference Paper %T SafeBimanual: Diffusion-based trajectory optimization for safe bimanual manipulation %A Haoyuan Deng %A Wenkai Guo %A Qianzhun Wang %A Zhenyu Wu %A Ziwei Wang %B Proceedings of The 9th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Joseph Lim %E Shuran Song %E Hae-Won Park %F pmlr-v305-deng25c %I PMLR %P 3218--3238 %U https://proceedings.mlr.press/v305/deng25c.html %V 305 %X Bimanual manipulation has been widely applied in household services and manufacturing, which enables the complex task completion with coordination requirements. Recent diffusion-based policy learning approaches have achieved promising performance in modeling action distributions for bimanual manipulation. However, they ignored the physical safety constraints of bimanual manipulation, which leads to the dangerous behaviors with damage to robots and objects. To this end, we propose a test-time trajectory optimization framework named SafeBimanual for any pre-trained diffusion-based bimanual manipulation policies, which imposes the safety constraints on bimanual actions to avoid dangerous robot behaviors with improved success rate. Specifically, we design diverse cost functions for safety constraints in different dual-arm cooperation patterns including avoidance of tearing objects and collision between arms and objects, which optimizes the manipulator trajectories with guided sampling of diffusion denoising process. Moreover, we employ a vision-language model (VLM) to schedule the cost functions by specifying keypoints and corresponding pairwise relationship, so that the optimal safety constraint is dynamically generated in the entire bimanual manipulation process. SafeBimanual demonstrates superiority on 8 simulated tasks in RoboTwin with a 11.1% increase in success rate and a 18.9% reduction in unsafe interactions over state-of-the-art diffusion-based methods. Extensive experiments on 4 real-world tasks further verify its practical value by improving the success rate by 32.5%.
APA
Deng, H., Guo, W., Wang, Q., Wu, Z. & Wang, Z.. (2025). SafeBimanual: Diffusion-based trajectory optimization for safe bimanual manipulation. Proceedings of The 9th Conference on Robot Learning, in Proceedings of Machine Learning Research 305:3218-3238 Available from https://proceedings.mlr.press/v305/deng25c.html.

Related Material