Sagitar Jutron’s second-generation all-solid-state sensing platform is released, providing a physical AI data portal

Author｜Huang Nan Editor｜Yuan Silai The current embodied intelligence track is completing a key transformation. In the past, a large number of R&D and demonstrations in the industry focused on stage demonstrations of humanoid and quadruped robots. Now, industry evaluation standards are based on long-term stable operation capabilities in real scenes such as factories, parks, and homes. But a real problem is that most basic models only rely on two-dimensional images and simulated virtual data for training, and lack native high-precision information about object distance, material, force, and dynamic interaction in the real three-dimensional space. As a result, robots can only understand pictures, but it is difficult to truly understand the operating rules of the physical world. Pure visual sensors are susceptible to interference from strong light, occlusion, rain and snow, and there are natural errors in ranging and spatial structure calculations. There is a huge gap between the physical rules of the simulation environment and the actual working conditions, which directly leads to a low success rate for tasks such as complex grabbing and long-sequence inspections after the robot leaves the exhibition hall, making it difficult to implement on a large scale. Among them, lidar is only regarded as a supporting navigation accessory for robots, responsible for a single obstacle avoidance function. Until the industry began to realize that high-precision three-dimensional perception is the core data source for the continuous iteration of physical AI. Sagitar Juchuang tried to make a different judgment. The Sagitar Jutron team pointed out that high-precision three-dimensional perception hardware is not only the eyes that allow robots to "see" the world, but also an indispensable source of data for the operation of physical AI. In order to open up a complete closed loop of "perception-understanding-decision-making-iteration", the generation logic of spatial data must be reconstructed from the underlying chip to continuously output real physical information that can be used for the continuous evolution of the model. From July 17th to 20th, the 2026 World Artificial Intelligence Conference (WAIC 2026) was held in Shanghai. RoboSense officially released the second-generation all-solid-state sensing platform E2 based on the self-developed "Peacock" SPAD-SoC chip. The system demonstrated the technical layout from the underlying chip to spatial sensing products. The second-generation all-solid-state sensing platform E2 (Source/Enterprise) The E2 platform is based on the self-developed ultra-large area array SPAD-SoC "Peacock" chip and 2D VCSEL chip. It adopts an all-solid-state architecture, and everything from signal transmission and reception to data processing is completed at the chip level. Compared with the previous generation of products, the E2 series has