deviceWISE Demonstrates Agentic AI Platform at IMTS 2026

deviceWISE, a Telit Cinterion company, will run three live production demonstrations of its deviceWISE platform at IMTS 2026, which will take place from September 14 to 19, 2026, in Chicago, Illinois. Exhibited at Booth 236475, the demonstrations will focus on visual inspection, robotic sorting, and automated assembly, demonstrating platform capabilities spanning asset connectivity, factory coordination, and edge AI deployment on active robotic cells.

Robotic Assembly and Automated Recovery Workflows
At the visual inspection cell, deviceWISE Visual Intelligence will examine circuit boards and additional physical components for missing parts and fabrication defects. A second station featuring agentic AI and digital twin integration will show a robot sorting colored cubes, while a third demonstration will use a FANUC robot to execute automated assembly of industrial components.

Both robotic installations will demonstrate Fault Detection & Recovery by intentionally triggering an operational fault and executing automated recovery through the platform. Traditional robotic line faults require manual identification, physical diagnostics, and operator intervention. The demonstrated system executes root-cause diagnostics directly at the edge, generating targeted recovery sequences and guiding operators through required resolution steps.

Intelligence Suite Architecture and Computer Vision Integration
The deviceWISE Intelligence Suite deploys industrial AI agents to process production telemetry and guide decision-making at the edge, targeting four functional operational areas:

Within the suite, Fault Detection diagnoses machinery and process deviations to execute or suggest corrective procedures. Concurrently, Workstation Sentinel monitors production line actions, validating live operations against defined standard operating procedures to identify misfed materials or incorrect components.

Scalable visual processing is integrated via the NVIDIA Metropolis Video Search and Summarization (VSS) Blueprint. Utilizing NVIDIA NIM microservices and the Model Context Protocol, the platform contextualizes real-time camera imagery with machine telematics and line process states. Visual events are evaluated against operational logic, allowing the system to route corrective actions to relevant plant systems or line personnel.

Additional Context
This section details technical specifications not included in the original news release.

Industrial edge IoT and data orchestration platforms interface directly with heterogeneous factory networks via native drivers for industrial protocols, including OPC UA, Modbus TCP, Siemens S7, Ethernet/IP, and SECS/GEM for semiconductor and electronics production. Edge AI processing on machine vision streams typically deploys on industrial PCs outfitted with high-performance edge accelerators running deep-learning architectures, such as convolutional neural networks (CNNs) and vision transformers (ViTs), processing high-definition camera feeds at frame rates between 30 and 60 frames per second. The Model Context Protocol (MCP) functions as an open standard enabling large language models and autonomous agentic workflows to interact directly with external data repositories, digital twin representations, and real-time execution layers. In robotic cells, deterministic edge nodes utilize standardized safety protocols (such as CIP Safety or PROFIsafe) and low-latency API connections to trigger programmable logic controller (PLC) routines and pause or redirect six-axis robot kinematic trajectories within response times under 20 milliseconds.

Edited by Romila DSilva, Induportals Editor, with AI assistance.