news

After Alibaba's investment, industrial capital increased its investment in humanoid robot company Zhuji Power

2024-07-15

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

On July 15, a reporter from Nandu learned that Zhuji Power, a humanoid robot startup, recently completed its Series A financing. Zhuji Power said that it is not convenient to disclose the specific amount of financing in the Series A for the time being.

In addition to Alibaba, which was disclosed in mid-May, the lead investors include China Merchants Venture Capital under China Merchants Group and Shangqi Capital, a private equity investment platform under SAIC Motor. Old shareholders in the angel and pre-A rounds, Frees Capital, Oasis Capital and Vision Capital also participated in the investment.

The three leading investors are industrial capitals, whose businesses cover multiple application scenarios such as AI, automobile manufacturing, logistics and urban integrated services. According to Zhuji Power, seeking support from industrial capital is related to the goal of this round of financing: to find matching strategic industrial resources and application scenarios, and to promote the verification and iteration of related technologies in real scenarios.

"We have locked in industrial investors from the very beginning in our Series A financing because scenarios represent time, money and technology." Zhang Wei, founder of Zhuji Power, said that this round of financing will help us to deepen our understanding of scenarios, speed up the construction of the basic large model of humanoid robots, and make technology iterations more in line with real needs.

The investment team of China Merchants Venture Capital stated that as a comprehensive central enterprise, China Merchants Group has rich application scenarios in the fields of transportation and logistics, comprehensive finance, and comprehensive development of cities and parks. The investment in Zhuji Power is an important step for the company in its industrial artificial intelligence innovation layout, and "we look forward to the implementation of general-purpose robots in our group's industries in the future."

The investment team of Shangqi Capital also looks forward to the commercialization of humanoid robots: "We look forward to working with Zhuji Power to accelerate the application of general humanoid robots in fields such as intelligent manufacturing."

For Alibaba, the investment in Zhuji Power means that Alibaba's AI big model investment map has further extended to the downstream of the industry chain. Alibaba has been active in investing in the field of big models over the past year. Behind the so-called "Five Little Tigers" of China's AI big models - Zhipu, Baichuan, MiniMax, Dark Side of the Moon, and Zero One Everything, Alibaba's investment can be seen. (See: Big Companies Compete for Humanoid Robots: Alibaba Invests in Startup Zhuji Power)

Shenzhen-based Zhuji Power was founded in 2022 by Zhang Wei, a professor in the field of robotics at the Southern University of Science and Technology. In November 2023, Zhang Li, COO of autonomous driving unicorn company WeRide, joined Zhuji Power as co-founder and COO.

Zhuji Power's product line includes full-size humanoid robots, four-wheeled and legged robots, bipedal robots and related solutions. Among them, the full-size humanoid robot CL-1 was announced to the public in December last year, and it can currently climb stairs one step at a time and run back and forth.


The movement picture of the Zhuji Power humanoid robot.

Zhuji Power introduced that the next research and development focus is to improve the motion intelligence of humanoid robots in real-world scenarios. In terms of implementation steps, first, a variety of human motion control data will be collected from sources such as Internet video data and human motion capture data; then pre-training will be carried out to improve the algorithm's basic motion understanding and generation capabilities, as well as multi-scenario generalization capabilities; finally, reinforcement learning will be used to narrow the gap between simulation data training results and real-world environment interactions.

Written by: Yang Liu, trainee reporter of Southern Metropolis Daily