Luxonis
US maker of OAK spatial AI cameras and DepthAI SDK, combining stereo depth sensing with on-device neural inference for robotics and vision applications.
Luxonis is a Denver, Colorado-based hardware and software company founded in 2019 by Brandon Gilles. The company specializes in spatial AI — the fusion of stereo depth perception and on-device neural network inference — delivered through its OpenCV AI Kit (OAK) camera product line and the open-source DepthAI SDK.
The company's founding story traces to an effort called Commute Guardian, a project aimed at protecting cyclists and pedestrians using real-time AI and depth perception. While building this system, the team discovered that no existing platform combined the right chip, hardware, firmware, and software in a single embedded form factor. To solve this, they created the DepthAI platform, which leverages Intel's Myriad X VPU to run neural network inference directly on the device alongside stereo depth estimation.
In 2020, Luxonis launched a Kickstarter campaign for OAK-1 (monocular AI camera) and OAK-D (stereo depth + AI), raising $1.3 million from over 6,500 backers in just 20 minutes — coinciding with OpenCV's 20th anniversary. Luxonis partnered with the official OpenCV.org organization to release the OAK hardware under an MIT-licensed open-source framework. A subsequent OAK-D Lite Kickstarter raised $1 million from 8,000 backers and funded in two minutes.
Since 2022, Luxonis has expanded its product catalog to include industrial PoE (Power over Ethernet) devices featuring rugged enclosures, scalable multi-camera configurations, and real-time AI processing for factory and outdoor deployment. The OAK-D Pro adds active stereo depth with IR dot projectors and night vision. Products are available through the Luxonis online store.
The DepthAI V3 SDK is the latest iteration of the company's open-source software API, enabling developers to write custom vision pipelines that run locally on OAK devices without cloud round-trips. The SDK supports Python and C++ and integrates with ROS.
Luxonis cameras serve applications in robotics, autonomous vehicles, drone navigation, industrial inspection, smart retail, and research. The company has approximately 87 employees as of mid-2026 and is headquartered at 1500 N Grant St, Suite 4217, Denver, Colorado 80203.
Primary type & automation activities this supplier delivers:
OAK-D
- Model: OAK-D
- Launch Year: 2020
Stereo depth + AI camera with on-device neural inference via Intel Myriad X VPU, open-source DepthAI SDK.
OAK-D Pro
- Model: OAK-D Pro
- Launch Year: 2022
Industrial-grade stereo AI camera with active IR stereo, night vision, and on-device neural inference.
Contact Luxonis
WEBSITE
https://luxonis.comHEADQUARTERS
United States
Company Facts
Founded
2019
Primary Role
Component Supplier
Company Size
-
Primary Region
North America
Annual Sales
-
Funding Stage
-
Funding Total
-
Tashan Technology Unveils Tactile Perception Loop for Enhanced Robotic Interaction Learning at WRC 2026
At the World Robotics Conference (WRC) 2026, tactile perception emerged as a focal point, highlighting its importance in robotic interaction. Tashan Technology showcased its tactile perception loop, addressing the limitations of visual imitation learning in complex industrial environments. The company aims to enhance robots' physical interaction capabilities by developing scalable tactile sensors that enable autonomous learning through real-time feedback. The significance of this development lies in its potential to overcome the challenges faced by robots in understanding physical interactions. Current reliance on visual data fails to capture the nuances of tactile feedback, leading to high failure rates in tasks involving unknown objects. Tashan Technology's approach seeks to fill this gap by providing robots with a tactile sensory system that allows them to independently comprehend and adapt to their physical surroundings. Looking ahead, Tashan Technology's advancements in tactile perception could revolutionize the robotics industry by enabling robots to perform tasks with greater accuracy and reliability. The company’s TS-F series sensors, capable of detecting various materials and providing high-resolution force feedback, represent a significant step towards achieving human-like tactile perception in robots. No further timeline was disclosed at the time of publication.
Tactile Perception Robotic Interaction Sensor Technology AI Robotics
Hanwei Technology Showcases Advanced Perception Solutions During 2026 World Robot Conference
During the 2026 World Robot Conference, Hanwei Technology Group Co., Ltd. showcased five types of embodied perception products, including tactile and olfactory sensors. The company also launched the HAU925, a nine-axis inertial sensor chip aimed at embodied intelligence, and demonstrated various advanced sensing technologies. Chairman Ren Hongjun discussed production rhythms and industry strategies, highlighting the growing importance of perception in robotics. The significance of these developments lies in Hanwei's focus on perception as a foundation for intelligent robotics. The company’s electronic skin technology, which has already reached over 30 clients, is poised to enter mass production, marking a critical point in its commercialization. Meanwhile, other sensory products are expected to scale up within two to three years, although significant profits may take longer to materialize. Looking ahead, Hanwei anticipates a shift in the robotics industry towards greater emphasis on embodied perception. However, challenges remain in integrating perception technologies into practical applications. The company aims to address these gaps by providing essential data collection infrastructure, including motion capture suits and data gloves, to support the industry's evolving needs.
Embodied Intelligence Sensor Technology Robotics AI Wearable Technology
Honor Robot Sets Three Human Records Using Suoteng Lidar Technology
Recently, at the Second World Humanoid Robot Sports Competition, the Honor humanoid robot 'Lightning' broke three records using Suoteng's lidar technology. It completed the 1500 meters, 400 meters, and 100 meters events in 2 minutes 30 seconds, 39.45 seconds, and 9.32 seconds, respectively, surpassing human world records in all three categories. This achievement highlights the capabilities of the 'Lightning' robot, which demonstrates stable perception abilities across different athletic conditions. The Suoteng lidar plays a crucial role in the robot's environmental perception system, continuously gathering three-dimensional spatial information to support high-speed movement and stable operation. As the Honor robot sets these records, Suoteng's robotics business is also accelerating its scale. In the first quarter of 2026, the company sold 185,500 lidar units in the robotics sector, marking a 1458.8% year-on-year increase. The company has established itself as a leader in global 3D lidar shipments for robotics and is poised for further growth with the upcoming mass production of its self-developed SPAD-SoC chips.
Humanoid Robots Lidar Technology Robotics AI Sports Technology
WRC 2026: A Sensor's Two Decades of Support for China's Robotics Sector
The 2026 World Robot Conference (WRC) commenced on August 19 in Beijing, showcasing over 300 exhibitors and more than 2,000 exhibits. Key players like Zhejiang humanoid NAVIAI WA2 and others highlighted the importance of 'embodied intelligence' through measurable physical metrics. The event emphasized the need for advanced sensors, with Yuli Instruments' new C-series six-dimensional force sensors gaining attention for their role in robotics. Yuli Instruments has established itself as a leader in sensor technology, having achieved UL certification and collaborated with ABB for over 15 years. Their products, which include over 1,000 types of sensors, are designed to withstand the demanding conditions of industrial applications. The company’s commitment to a fully autonomous production system and international testing standards has garnered trust from global clients, making it a key player in the robotics supply chain. As the industry evolves, the focus is shifting towards force perception as a critical factor for operational success in robotics. The ability to maintain stability on uneven surfaces and perform delicate tasks without damaging components is becoming essential. The increasing reliance on Yuli Instruments' sensors within the ABB ecosystem indicates a growing recognition of their importance in achieving reliable robotic performance.
Robotics Sensor Technology Industrial Automation Humanoid Robots
Huaweike Unveils Advanced Tactile Sensors at WRC 2026 to Enhance Robotics
On August 19, 2026, the World Robot Conference opened in Beijing, where Wuhan Huaweike Intelligent Technology Co., Ltd. launched its Wanxiang series of tactile sensors. The event highlighted the industry's growing maturity in algorithms and visual perception, while the lack of tactile data remains a significant barrier to practical applications in robotics. The advancements in embodied intelligence over the past two years have been notable, with humanoid robots achieving over 20 degrees of freedom and costs reduced to under 100,000 yuan. However, challenges persist in fine manipulation tasks, particularly in the 'grasping' action, where robots often fail due to insufficient tactile feedback. Traditional sensors typically have an accuracy of 5% FS, which is inadequate for delicate operations. The Wanxiang series boasts an impressive accuracy of 1% FS and a minimum trigger force of 0.01N, setting a new standard for tactile data quality. This enhancement allows for the detection of minute weight changes, significantly improving the granularity of data collected. Huaweike's data collection gloves and simulation tools aim to address the efficiency of large-scale tactile data acquisition, reducing costs by 90% and increasing efficiency tenfold, which is crucial for training advanced robotic models.
Tactile Sensors Data Collection Robotics AI Industrial Automation
Wutong Introduces THD1000R Gloves for Advanced Tactile Data Collection in Various Settings
Wutong has launched the THD1000R, a groundbreaking solution for tactile data collection in home, industrial, and laboratory environments. This innovative glove utilizes wireless capacitive sensing technology to capture intricate tactile signals in complex operational settings, marking a significant advancement in the field of haptic perception. The introduction of the THD1000R is crucial as the industry transitions towards real-world data applications, moving beyond mere hardware demonstrations. With advancements in language and vision models, the need for enhanced tactile data collection capabilities becomes evident, especially as the sector currently faces challenges in engineering viable products. Looking ahead, the THD1000R's ability to provide high-precision data collection will be vital for applications requiring nuanced physical interactions. No further timeline was disclosed at the time of publication, but the glove's performance in capturing subtle tactile signals will be essential for effective real-world applications and the development of advanced robotic systems.
Tactile Sensors Data Collection Wearable Technology Robotics Embodied Intelligence
Tsinghua University Team Unveils Soccer-Playing Humanoid Robot in Science Robotics
On August 19, a team led by Professor Zhao Mingguo from Tsinghua University's Department of Automation published a paper in Science Robotics titled "Learning Vision-Driven Reactive Soccer Skills for Humanoid Robots." The research introduces a unified perception-motion reinforcement learning framework that enables a humanoid robot, known as Booster, to autonomously locate and kick a soccer ball without relying on external positioning or motion capture systems. This development is significant as it addresses the challenges of dynamic environments where visual perception is often unreliable. Traditional robotic soccer systems typically follow a modular pipeline, which can amplify errors due to perception noise and delays. The new framework allows the robot to make accurate movements even under imperfect visual conditions, marking a notable advancement in the field of robotics. Looking ahead, the implications of this research could extend to various applications in robotics, particularly in environments where visual feedback is compromised. The paper's publication in a prestigious journal underscores its importance, and the team’s innovative approach may pave the way for future developments in humanoid robotics and autonomous systems. No further timeline was disclosed at the time of publication.
Humanoid Robots Reinforcement Learning Robotics Research AI Autonomous Systems
Humanoid Robot Crashes During Training for World Humanoid Robot Games in Beijing
A humanoid robot experienced a severe crash during a training sprint in Beijing, colliding with a cushioned wall and nearly splitting in half. This incident occurred as part of the preparations for the World Humanoid Robot Games, also known as the 'Robot Olympics,' set to begin on August 22. The event will feature 2,056 machines from 666 teams across 16 countries. The crash, which was captured on video and has garnered over 13 million views online, raises questions about the robot's programming and perception capabilities. Experts note that such failures are crucial for testing robots under challenging conditions, allowing developers to gather data to enhance performance and safety before real-world deployment. As the World Humanoid Robot Games approaches, with 30 competitive events planned, the increase in participation—up 138 percent in teams and quadrupled robot entries—highlights the growing interest and advancements in humanoid robotics. No further timeline was disclosed at the time of publication.
AI and Robotics
Innovative Hybrid Robot Skin Integrates EIT Mapping with Pneumatic Force Sensing
A new prototype hybrid robot skin has been developed, integrating electrical impedance tomography (EIT) location mapping with pneumatic force sensing. This innovative design features a rigid base, four soft air-tight pads, a continuous piezoresistive layer, and 32 boundary electrodes, all produced through advanced 3D printing and spray coating techniques. This development is significant as it enhances the robot's ability to perceive its environment and respond to physical interactions, which is crucial for applications in humanoid robotics and other fields. The combination of EIT and pneumatic sensing allows for more precise location mapping and force detection, potentially improving the functionality and safety of robotic systems in various environments. Looking ahead, the implications of this technology could lead to more adaptive and responsive robotic systems. As researchers continue to refine this hybrid skin, it will be important to monitor advancements in its application across different robotic platforms and industries. No further timeline was disclosed at the time of publication.
Fourier Unveils GR-3C Robot with RoboSense LiDAR at WRC 2026
During the 2026 World Robot Conference (WRC), Fourier showcased its third-generation humanoid robot series GR-3 at the Beijing Yichuang International Exhibition Center. The series includes the GR-3, focused on interactive services, and the GR-3C, which emphasizes functional services. The GR-3C 'Astronaut' Ultra version features RoboSense's first hemispherical digital LiDAR, Airy, designed for embodied robots. The Airy LiDAR employs a 'SPAD-SoC+VCSEL' digital chip architecture, providing a compact size similar to a ping pong ball while delivering a 360°×90° ultra-wide field of view, a 60-meter radius coverage, and centimeter-level detection accuracy. This technology allows for flexible integration into various parts of the robot, continuously gathering extensive and precise 3D environmental data, essential for reliable perception and navigation in complex scenarios. At the exhibition, Fourier presented its self-developed 'Embodied Home' technology solution, creating an immersive experience titled 'A Day in the Life of a Robot Butler.' Multiple GR-3 robots acted as intelligent assistants in simulated home environments, responding to natural language commands to autonomously perform tasks such as home security, item delivery, companionship, and guest reception. The interaction evolved from simple command execution to understanding and proactive service for users.
Humanoid Robots LiDAR Technology Robotics AI Home Automation
ROBOTTODAY Weekly June 30 – July 4, 2026
Unitree wins CSRC approval for China’s first humanoid IPO, Korea unveils ₩312T AI investment, RoboCup 2026 opens in Incheon, and Paris launches Europe’s inaugural Physical AI summit.
