Gasgoo Munich-QJ Robots has recently closed an A+ funding round worth hundreds of millions of yuan, according to Gasgoo Embodied Intelligence. The round drew participation from prominent market institutions and strategic investors, including Yuanhe Houwang, Xiaoguang Capital, Xineng Venture Capital, Inno Angel Fund, Jingming Capital, and Future Marginal Ventures.Proceeds will primarily fuel core technology R&D, engineering capabilities, and expansion into real-world scenarios. The goal is to deploy Qianjue’s robotic intelligence across a wider range of form factors and use cases.Image credit: QJ RobotsQJ Robots focuses on real-world robotic intelligence. Built on a predictive world model and a polynomial representation architecture, the company is constructing infrastructure that spans perception, decision-making, planning, and control. Designed to adapt to various robot forms, this system helps manufacturers cut redundant R&D and accelerate product launches and scale.The predictive world model does not aim to reconstruct every pixel and detail of reality. Instead, it zeroes in on state changes that directly impact task completion, forecasting potential outcomes within a feature space.This approach allows robots to link their current state with action results and subsequent decisions. During execution, if people, objects, or the environment shift, the system updates its assessment based on new observations and re-plans next steps.Meanwhile, the polynomial representation architecture is engineered to boost training and adaptation efficiency. It effectively expresses and combines key features that drive task execution, supports simplicity regularization in the representation space, and mitigates the risk of representation collapse. This focused approach ensures the model prioritizes information that is genuinely relevant to the task.Leveraging this architecture, Qianjue can more effectively utilize multi-source data—such as first-person and human operation videos—improving sample efficiency and reducing reliance on single data sources for training robotic capabilities.For robots of varying forms, the system tailors capabilities by trimming, combining, and reinforcing them based on sensor configurations, actuator structures, and performance limits. This enables a single set of intelligence capabilities to adapt across different morphologies.Currently, Qianjue is validating and adapting its intelligence for a diverse range of forms, including humanoids, wheeled dual-arm and single-arm robots, quadrupeds with arms, cleaning robots with arms, and cantilever drones. Applications are steadily rolling out across home, retail, and service settings.About "Seeds Discovery":Gasgoo’s "Seeds Discovery" column is designed as a service platform connecting startups, ecosystem partners, investors, and local governments to deeply empower the automotive supply chain. Since its inception, the column has sought to identify standout companies, technologies, and business models that inspire and lead the industry amid the intelligent transformation, driving the growth of automotive innovation. According to Gasgoo, nearly every startup featured in "Seeds Discovery" has successfully connected with ecosystem resources across the supply chain.