Bangjun Wang

RoboVerse is open-sourced now!🚀 It's the most ambitious work I've ever participated in👏. Check it out at roboverse.wiki 🔥

Still keeping crawl at half-speed but it works with the costume on!

🏆 Announcing #IROS RoboGen Best Open Source Award: RoboVerse @FeishiWang @siyuanhuang95 @yuewang314

Ever want to enjoy all the privileged information in sim while seamlessly transferring to the real world? How can we correct policy mistakes after deployment? 👉Introducing GSWorld, a real2sim2real photo-realistic simulator with interaction physics with fully open-sourced code.

In my past research experience, finding or developing an appropriate simulation environment, dataset, and benchmark has always been a challenge. Missing features, limited support, or unexpected bugs often occupied my days and nights. Moreover, current simulation platforms are relatively fragmented—making it challenging to replicate the success of the RT-X dataset in unifying community efforts. Introducing RoboVerse, we provide a unified platform, dataset, and benchmark for scalable and generalizable robot learning. We hope to build a shared foundation to combine the community efforts. RoboVerse includes: MetaSim: We carefully designed a configuration system and a universal interface to align current robotic simulators. With MetaSim, you can use any simulator with the same code—bringing together the community’s diverse efforts under one framework! RoboVerse Dataset and Benchmark: We unify popular simulation environments and benchmarks into a single cohesive system and introduce the RoboVerse dataset—a large-scale, high-quality synthetic dataset. Additionally, we propose a standardized benchmark across both imitation learning and reinforcement learning. A cool feature enabled by our unified framework: Hybrid Simulation! You can now integrate physics engines and renderers from different simulators—e.g., using MuJoCo precise physics with Isaac photorealistic rendering. This not only elevates simulation fidelity but also significantly enhances real-world transfer performance across complex robotic applications. Hopefully, our team’s efforts could serve the robotic community to thrive vibrantly in the years to come. RoboVerse is open-sourced🥳!!! Project Page: roboverseorg.github.io Documentation: roboverse.wiki Github Repo: github.com/RoboVerseOrg/R… Paper: roboverseorg.github.io/static/pdfs/ro…

🚀 RoboVerse has been accepted to RSS 2025 and is now live on arXiv: arxiv.org/abs/2504.18904 ✨ Also be selected in HuggingFace Daily: huggingface.co/papers/2504.18… 🛠️ Explore our open-source repo: github.com/RoboVerseOrg/R… We're actively developing and adding new features daily — come explore with us and enjoy!

Really cool and ambitious work. Robotics simulation is really challenging and fragmented, and it's really possible that the GPT moment for robotics can really only come when we have our MMLU. And that means simulations, because we need something comparable and reproducible.

First milestone reached. It’s a small step, but it brings me a bit of comfort today.🤞🏻

analyzing @BostonDynamics new demo [Atlas is autonomously moving engine covers between supplier containers and a mobile sequencing dolly. The robot receives as input a list of bin locations to move parts between.] youtube.com/watch?v=F_7IPm… [1/n]

We’ve seen humanoid robots walk around for a while, but when will they actually help with useful tasks in daily life? The challenge here is the diversity and complexity of real-world scenes. Our new work tackles this problem via 3D visuomotor policy learning. Using data from only 1 scene, our Improved 3D Diffusion Policy (iDP3) enables a full-sized humanoid robot to autonomously pick&place objects, pour water, and wipe tables, in the wild open world. (and all these skills are useful, right?) Web: humanoid-manipulation.github.io Fully open-sourced code: github.com/YanjieZe/Impro…

@ZeYanjie Congrats Yanjie!👏

Happy to see 3D Diffusion Policy (DP3, 3d-diffusion-policy.github.io) is selected into State of AI Report. Thank you Nathan! DP3 is our initial trial on universal 3D visuomotor policy learning. With DP3, you can demonstrate lots of tasks, on robot arms, legged dogs, even humanoids. 1/n

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”

We fully open-sourced our humanoid learning & deployment pipelines for both H2O (IROS'24) and OmniH2O (CoRL'24). This includes simulation training, sim2real, hardware setup, and deployment codes. Repo: github.com/LeCAR-Lab/huma… Led by @TairanHe99 @zhengyiluo

Cool application leveraging our recent real2sim evaluation project SIMPLER (simpler-env.github.io) for an LMSys like robot evaluation platform 🚀

Want to scale RL with your shiny new GPU? 🚀 In our ICML24 Oral we find that RL algorithms hit a barrier when data is scaled up. Our new algorithm, SAPG, proposes a simple fix. It scales to 25k envs and solves hard tasks where PPO makes no progress. sapg-rl.github.io 1/n

Mobile robots and manipulation! 🕹️ Cooperative mobile manipulation is a crucial area in robotics, focusing on enabling robots to work together on tasks that require collaboration, much like humans do. This field addresses challenges such as transporting heavy or unwieldy objects in complex environments. To address these challenges, researchers from the Computational Robotics Lab - ETH Zürich have developed a multi-robot, bi-level optimization system based on direct transcription for wheeled mobile manipulation. This innovative approach utilizes static forces calculated for stability objectives at a lower level to inform wheeled trajectory planning at a higher level, resulting in effective planning and safer execution. Additionally, this system has been integrated with a previously developed Mixed-Reality interface! Congrats to authors @FloMaushart! Tech stack and partners: @Universal_Robot, @clearpathrobots, @Microsoft & @Hiltigroup! 🦾 ~~~ ♻️ RT to help 1 robot find a new workplace.

Just wrapped up my incredible #RSS2024 journey in Delft 🇳🇱. This was my first in-person conference! Thrilled to have made so many new friends and talked directly with those top robotics researchers during the conference ! 👍 Really looking forward to meet you all again!

@ChongZitaZhang @RoboticsSciSys @TairanHe99 @_wenlixiao @guanqi_he @GuanyaShi @maththrills Congrats！

Very honoured to receive this finalist certificate at #RSS2024 @RoboticsSciSys Thanks to all my amazing coauthors @TairanHe99 @_wenlixiao @guanqi_he Changliu @GuanyaShi, the great reviewers and AC who helped us a lot in improving the paper, and the committee @maththrills et al

Arrived in Netherlands but temporarily lived in Amsterdam today 🇳🇱 Excited to meet with you in Delft!