This technology is used by autonomous mobile robots (AMRs), quadrupeds, humanoid robots, and fixed-arm perception-guided industrial robotics. The cameras use an on-board application-specific integrated circuit (ASIC) to calculate geometric distance while combining two imaging sensors to capture a scene from different angles. RealSense creates advanced computer vision technology that helps machines see, understand, and interact with the world.
It includes 30+ security capabilities2 for comprehensive protection across the entire compute stack. A wall that reacts when a child bounces a ball off it, a projected room that a shopper can walk through, a museum floor that responds to footsteps — all of it depends on a system that can see people and space in real time, at human scale, without asking anyone to strap on hardware. RealSense stereo depth cameras give robotic arms, cobots, and the systems watching over them a continuous, geometry-based understanding of the floor, so automation adapts in real time instead of stopping the line when something moves. RealSense closes that gap with stereo depth cameras that deliver millimeter- and sub-millimeter-level dimensional accuracy, global-shutter capture for objects in motion, RealSense Spins Out of Intel as 3D Perception Demand Grows and an open-source SDK that turns raw depth data into production-ready measurement, volume, and dimensional-weighing tools. The advanced movement and navigation system was developed using NVIDIA Isaac Lab, a highly realistic virtual testing environment. With improved path planning, robots can quickly adapt to changes in their environment, such as moving objects or someone stepping into their path.
- Its recently launched D555 depth camera, powered by the next-gen RealSense Vision SoC V5 and featuring Power over Ethernet (PoE), demonstrates the company’s ongoing leadership in embedded vision technology and edge AI capabilities.
- RealSense has developed robust, global manufacturing technology capabilities to ensure consistent quality and product performance, working with a broad network of vision system distributors and value-added resellers.
- Built for advanced Physical AI systems, the RealSense D450 Series extends the proven D430 architecture to longer-range, rugged deployments.
- For longer range applications, the D455 features wide field of view and global shutter on the depth and RGB sensors.
- The D436 combines the D430 wide-FOV depth module with a synchronized, global-shutter RGB sensor whose field of view closely matches the depth sensors.
The development of this sophisticated locomotion and navigation stack was accelerated using NVIDIA Isaac Lab, which served as a high-fidelity digital proving ground for reinforcement learning and policy training. About CognexFor over 40 years, Cognex has been making advanced artificial intelligence machine vision easy, paving the way for manufacturing and distribution companies to become faster, smarter, and more efficient through automation. The combination of a wide field of view and global shutter sensor on the D435i make it the preferred solution for applications such as robotic navigation and object recognition. An IMU combines a variety of sensors with gyroscopes to detect both rotation and movement in 3 axes, as well as pitch, yaw and roll. The acquisition represents a significant expansion of Cognex’s technology portfolio as the company positions itself to serve both established industrial automation markets and emerging robotics applications.
Cognex Set to Buy RealSense for Robot Vision Tech – Design News
Cognex Set to Buy RealSense for Robot Vision Tech.
Posted: Mon, 05 Oct 2026 19:51:05 GMT source
Renishaw Introduces LIBERTAS Software to Reduce Support Structures in Metal AM
- With improved path planning, robots can quickly adapt to changes in their environment, such as moving objects or someone stepping into their path.
- The advanced movement and navigation system was developed using NVIDIA Isaac Lab, a highly realistic virtual testing environment.
- It must function as the robot’s visual cortex, enabling accurate localization, collision avoidance, terrain understanding, and stable, predictable motion in unstructured environments,” said Nadav Orbach, CEO of RealSense, in a statement.
- These include climbing stairs, detecting curbs and height changes, walking on uneven ground, and avoiding moving obstacles such as people or equipment.
Vision is a key element to enable this awareness.Depth information takes this awareness to the next level and opens the door to even more use cases and powers applications such as navigation, object detection, distance or volume measurements or human-machine interactions in industrial, commercial and consumer devices. Originally founded by Intel in 2014 and spun out in 2025, RealSense has established a leading position in 3D robotic perception, with applications across fixed arm perception-guided robotics, autonomous mobile robots, quadrupeds and humanoids. A cross-platform, open-source software stack called the RealSense SDK (formerly housed as the librealsense repository on GitHub) is used to machine learning integrate RealSense hardware.
- Stereo cameras can see further but accuracy degrades with distance and varies depending on scene and lighting conditions.
- Powered by the RealSense™ Vision Processor D4, the D430 Series performs depth processing onboard and delivers high-quality 3D data across robotics, industrial automation, inspection, analysis and other vision applications.
- Vision is a key element to enable this awareness.Depth information takes this awareness to the next level and opens the door to even more use cases and powers applications such as navigation, object detection, distance or volume measurements or human-machine interactions in industrial, commercial and consumer devices.
- Originally founded by Intel in 2014 and spun out in 2025, RealSense has established a leading position in 3D robotic perception, with applications across fixed arm perception-guided robotics, autonomous mobile robots, quadrupeds and humanoids.
- Global-shutter stereo imagers capture an entire frame simultaneously, making the D430 architecture well suited to moving robots, moving objects and rapidly changing viewpoints.
- A cross-platform, open-source software stack called the RealSense SDK (formerly housed as the librealsense repository on GitHub) is used to integrate RealSense hardware.
- With ruggedized enclosures, secure interfaces, and extended I/O capabilities, these 3D depth cameras are designed for logistics, factory automation, robotics, and other demanding applications.
These include climbing stairs, detecting curbs and height changes, walking on uneven ground, and avoiding moving obstacles such as people or equipment. It must function as the robot’s visual cortex, enabling accurate localization, collision avoidance, terrain understanding, and stable, predictable motion in unstructured environments,” said Nadav Orbach, CEO of RealSense, in a statement. Using NVIDIA Isaac Lab, LimX Dynamics trained humanoids in simulation, bridging sim-to-real gaps and ensuring safe, reliable 3D navigation before debut. RealSense delivers https://aicurated.com/ industry-leading depth cameras and vision technology used in autonomous mobile and humanoid robots, access control, industrial automation, healthcare and more. It must function as the robot’s visual cortex, enabling accurate localization, collision avoidance, terrain understanding and stable, predictable motion in unstructured environments.” “Humanoids operate in three dimensions, alongside people, in environments that are constantly changing,” said Nadav Orbach, CEO of RealSense.
Twice the Range for Demanding Physical AI Applications
Cognex expects the combination of its existing machine vision capabilities and RealSense’s 3D perception platform to provide customers with a more comprehensive range of visual intelligence technologies. The acquisition also supports Cognex’s strategy of diversifying its revenue growth by expanding large language models into adjacent technology markets. The company’s products help manufacturers and warehouse operators automate inspection, identification, measurement and other processes. RealSense’s depth-sensing technology is designed to provide the visual information needed for those applications.