Tech distributor Avnet and robotics integrator Weston Robot are bringing edge-based artificial intelligence to the factory floor. The companies are creating a new real-time autonomous inspection platform powered by quadruped robots.
Weston Robot integrates quadruped, humanoid, and mobile robot-based applications for its customers. Applications range from inspection to security and facility management. The company primarily resells Unitree robotic solutions. It also manufactures custom autonomous mobile robot (AMR) platforms and unmanned maritime systems. The Singapore-based company primarily offers the solutions in a robots-as-a-service (RaaS) model.
Avnet is a supplier of industrial compute architecture. For the latest inspection platform, Avnet attached a backpack filled with Avnet components to a quadruped. The Avnet architecture delivers up to 50 TOPS of AI performance. It does this while supporting AI-driven inspection, thermal and visual analytics, 3D lidar mapping, and dependable operation in GPS-denied environments.
Powered by AMD Ryzen AI Embedded processors, the platform executes AI workloads directly on the device. It enables low-latency AI inference, faster decision-making, and reliable operation even where cloud connectivity is limited. By processing data at the edge, organizations can detect operational issues earlier, reduce inspection costs and improve operational continuity.
“Physical AI represents the next evolution of artificial intelligence—one where intelligent systems can perceive, reason and act autonomously in the real world,” said Arthur Chung, Vice President, Sales & Supplier Management, Avnet Asia. “Through our collaboration with Weston Robot, we are bringing together advanced computing, edge AI and robotics to help organizations scale industrial intelligence, strengthen safety and improve operational resilience. As organizations move beyond AI experimentation, intelligent autonomous systems will become an essential part of smarter, safer and more efficient industrial operations.”
Combined system will be more flexible than previous systems #
When combined with Weston’s fleet management software, the robot can be set on a mission to patrol and inspect a defined path through a facility. Using onboard cameras and other sensors, the robot can stop and investigate any number of inspection sites along its patrol route.
Using the onboard Avnet computing architecture, new inspection and recognition tasks can be trained and deployed to the robot. The robot leverages AI for new inspection or perception tasks. Avnet enhances the perception capabilities of the underlying quadruped robot. The company said it enables new capabilities not possible with the basic Unitree quadruped robot. “Industrial environments generate enormous volumes of operational data, yet many inspection processes remain manual, reactive and resource-intensive,” said Dr. Zhang Yanliang, Chief Scientist, Weston Robot. “By combining robotics, AI and edge computing, we are enabling a new generation of intelligent systems that continuously monitor critical infrastructure, detect anomalies earlier and deliver actionable insights in real time. We believe Physical AI will become a foundational technology powering the next generation of industrial operations.”