en.Wedoany.com Reported - Orbbec is showcasing its latest 3D vision products and solutions at Automate 2026 (North Hall, Booth 14045), aiming to accelerate industrial automation. To meet the growing industry demand for perception-driven intelligent automation and robotics, the company has launched a portfolio of industrial-grade 3D vision cameras for complex scenarios, along with integrated AI solutions.

With the convergence of industrial vision and edge AI, Orbbec is demonstrating a hardware-software integrated solution that combines high-precision 3D vision hardware with AI models. This solution is designed to enhance robots' spatial perception and environmental understanding to address sensing challenges in complex industrial scenarios. To overcome automation bottlenecks encountered by traditional 3D vision hardware with transparent objects, low-texture or repetitive pattern surfaces, and highly reflective materials, Orbbec has partnered with Robbyant, an affiliate of Ant Group, to introduce the LingBot-Depth (LingBot Enhanced Depth Filter) for the Gemini 330 series. This filter is trained using chip-level high-precision data from the Gemini 330 series as standard input and is integrated with Robbyant's Vision-Language-Action (VLA) model. Robbyant has indicated that incorporating high-quality depth data directly into large models can improve robot manipulation capabilities and success rates. The filter is compatible with passive stereo input and achieves optimal performance when using active stereo input; it is fully compatible with the NVIDIA Jetson Orin platform, can be extended via the Orbbec SDK, and leverages NVIDIA CUDA and TensorRT for accelerated AI inference, enabling real-time edge inference for depth hole filling, noise removal, and edge refinement. Tasks supported by this solution include carton edge tracking in logistics, reflective metal alignment in assembly, and transparent container handling in laboratory automation.

Addressing industrial challenges such as precision assembly, dynamic warehousing, and high-vibration environments, Orbbec is showcasing a series of live demonstrations. Its customizable 3D DLP structured light camera achieves approximately 0.07 mm VDI/VDE accuracy and 0.05 mm Z repeatability at a distance of 0.5 meters. Featuring a compact and lightweight form factor, the camera seamlessly integrates with robotic arms, outputs high-resolution RGB data to support AI-driven object detection and deep learning, and offers adaptive multi-mode acquisition including fast, high-reflection, and precision modes.

For industrial logistics scenarios, Orbbec is presenting a forklift automation solution powered by the industrial-grade stereo 3D camera Gemini 435Le. This solution provides low-noise, high-density depth data in dynamic environments, supporting localization, navigation, and obstacle avoidance for detecting forklift tines and ground obstacles. Additionally, Orbbec is showcasing the ultra-compact stereo 3D camera Gemini 305g, designed for robot wrist mounting, fully compatible with the NVIDIA Jetson Orin platform, featuring GMSL2/FAKRA connectivity and low-loss cables to withstand mechanical vibration and electromagnetic interference.

Orbbec's 3D vision solutions are deployed through collaborations with global automation companies. Teradyne Robotics uses the Orbbec Gemini 305g stereo 3D camera in its UR AI Trainer. The UR AI Trainer is a platform for data collection required for training AI and VLA models, helping to handle operational tasks such as electronic assembly.

Orbbec has served over 1,600 robotics companies worldwide, achieving more than 70% market share in both the Chinese service robot vision market (source: GGII) and the South Korean mobile robot vision market (source: Interact Analysis). The company's manufacturing footprint includes an intelligent manufacturing base in Shunde (Foshan) and the upcoming RVMC factory in Vietnam. Through its multi-site manufacturing strategy, the company's annual manufacturing capacity exceeds 6 million robots.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com