Ropedia Secures $30 Mn to Scale Data Infrastructure for Physical AI

Funds have been raised in two pre-A rounds and can again international enlargement of the corporate’s real-world interplay knowledge platform for robotics and embodied AI programs

Ropedia, a Singapore-based startup constructing knowledge infrastructure for bodily AI, right this moment introduced it has raised US$30 million {dollars} in pre-A funding from enterprise traders with deep expertise in AI, enterprise know-how and infrastructure throughout Southeast Asia. Other funding members embody long-term monetary traders and strategic companions working in fields resembling robotics, mobility and enterprise deployment. Funding will scale Ropedia’s assortment of real-world, multimodal interplay knowledge and increase supply to know-how firms creating robots and embodied AI programs.

“A robotic can’t play baseball by watching a video any greater than you may be taught to journey a motorcycle by studying about it. The robotic should perceive what it’s like to grip a bat and know the timing it takes to hit a ball. That’s the data Ropedia’s know-how supplies, and it’s why this funding issues. Text scraped from the web was used to prepare the final era of AI. Real-world human expertise, captured on the identical scale, will prepare bodily AI. Physical AI will allow us to go away the lab and go to work, first in factories, then at dwelling, serving to our households,” stated Zhaoxi Chen, chief govt and co-founder of Ropedia.

From real-world sensing to model-ready intelligence, bodily AI requires an end-to-end knowledge pipeline. Ropedia integrates {hardware}, knowledge processing and knowledge supply into one scalable ecosystem. HOMIE, Ropedia’s wearable, head-mounted system, captures first-person video, audio, depth, hand monitoring, gaze, physique movement and digital camera pose concurrently, with every stream timestamped. This exact alignment issues as a result of bodily AI is dependent upon correlating perceptions and actions in actual time, a functionality most present knowledge assortment strategies can’t present. HOMIE feeds this synchronized knowledge into Ropedia’s processing and annotation programs.

Ropedia is an information infrastructure firm, not a data-labeling service. Standard labeling suppliers annotate knowledge that already exists; Ropedia generates the info itself, then synchronizes, buildings and regularly refines it end-to-end. This holistic strategy additionally units HOMIE aside from teleoperation-based knowledge assortment, which depends on bodily robotic {hardware} and is often restricted to particular robotic sorts.

The firm operates a “closed-loop pipeline” that integrates multimodal synchronization and rigorous high quality assurance into dataset era and model-aligned fine-tuning. It sells that pipeline in 3 ways: dataset licensing, selective entry to HOMIE {hardware} and analysis collaboration. HOMIE might be deployed anyplace and worn by anybody, giving Ropedia a scalable approach to accumulate synchronized, multimodal knowledge in parallel throughout many environments and customers. Unlike teleoperation-based programs, that are constrained by costly robotic fleets, the platform scales just by including wearable seize units.

This infrastructure powers Xperience-10M, one of many largest human-experience datasets within the business: 10 million interplay episodes and greater than 10,000 hours of multimodal recordings, spanning billions of synchronized video, depth, motion-capture and inertial-sensor frames. Each new HOMIE deployment additionally expands the vary of environments, behaviors and interactions the dataset captures, making it extra beneficial because the community grows.

Ropedia’s strategy cuts data-collection prices by up to 50 occasions in contrast with conventional strategies, and HOMIE has entered mass manufacturing. Ropedia has served greater than a dozen North American firms within the fields of embodied AI and spatial intelligence.

Pre-A fundraising was in two rounds and backed by particular person angel traders. The newest spherical, introduced right this moment, raised US$22 million and follows an earlier US$8 million spherical introduced on social media on March 16, for a complete of US$30 million. Ropedia will direct the funding towards three priorities: increasing knowledge assortment throughout Southeast Asia and North America, scaling {hardware} deployment to help bigger fleets of HOMIE units, and advancing its AI analysis and knowledge platform work, together with new engineering hires within the U.S.

One of its angel traders, a analysis scientist from Amazon, stated, “I backed the Ropedia workforce early as a result of they’d a uncommon mixture of deep technical experience, pace of execution and a transparent imaginative and prescient for the place bodily AI was heading. Since then, they’ve constructed a compelling knowledge infrastructure platform serving main robotics and foundation-model firms globally. I consider Ropedia is nicely positioned to turn into a foundational firm within the bodily AI ecosystem.”

Robotics and bodily AI lab prospects get extra knowledge, delivered quicker, throughout a wider vary of environments and duties. Ropedia serves as an information infrastructure layer for bodily AI, comparable to the info facilities that underpinned cloud computing or the web textual content used to prepare language fashions.

Chen co-founded Ropedia with Fangzhou Hong, chief know-how officer, and Ziwei Liu, chief scientist and an affiliate professor at Nanyang Technological University in Singapore. Chen is understood for his pioneering work in 3D laptop imaginative and prescient and multimodal AI, whereas Hong beforehand labored on Meta’s selfish multimodal intelligence analysis earlier than contributing foundational analysis in 3D spatial intelligence.

The publish Ropedia Secures $30 Mn to Scale Data Infrastructure for Physical AI first appeared on AI-Tech Park.