Why it matters now: The boundary between deterministic PLC control and probabilistic AI inference is dissolving faster than most factory operators realize. As industrial edge AI moves from pilot programs to production floors, the alliance between AI chipmaker Mobilint, industrial computing veteran ADLINK, and rugged-hardware specialist Getac signals a structural shift in how intelligence gets embedded at the machine level — not in the cloud, but right alongside the PLC.
Inside the Mobilint–ADLINK–Getac Partnership
Mobilint announced a trilateral collaboration with Taiwan-based ADLINK Technology and Getac to co-develop edge AI solutions purpose-built for industrial environments. Under the agreement, Mobilint contributes its proprietary neural processing unit (NPU) silicon and a full-stack software development kit (SDK), while ADLINK — a long-established player in industrial automation hardware that routinely interfaces with PLC systems — leads integration and deployment architecture.
Getac brings its expertise in ruggedized computing, ensuring the resulting platforms withstand the thermal stress, vibration, and electromagnetic interference common on factory floors. The combined offering targets a persistent industry pain point: real-time AI inference that coexists with, rather than replaces, the deterministic logic of traditional PLC controllers.
Analyst Insight: The Coexistence Model
Industry analysts note that the "AI alongside PLC" paradigm — rather than "AI replacing PLC" — represents the most commercially viable near-term path. PLCs handle safety-critical, deterministic tasks with microsecond precision, while NPU-driven edge AI augments them with anomaly detection, predictive maintenance, and visual quality inspection. This architectural separation preserves IEC 61131-2 compliance while unlocking new intelligence layers.
Edge AI and PLC: A Converging Landscape
The Mobilint deal did not emerge in isolation. At COMPUTEX 2026, the company signed separate memoranda of understanding with DFI, NEXCOM's NEXCOBOT division, and TOP MEASURE — all industrial computing or robotics specialists that routinely embed their hardware alongside PLC-based control architectures. The pattern is unmistakable: AI inference silicon vendors are racing to secure integration partnerships across the industrial automation supply chain.
This convergence reflects a broader market reality. The industrial edge market — spanning edge devices, compute units, servers, and networking gear — is projected to more than double from USD 21.19 billion in 2025 to USD 44.73 billion by 2030, according to MarketsandMarkets data. Within that, factory-floor edge AI industrial PCs alone are forecast to grow from USD 0.68 billion in 2026 to USD 1.37 billion by 2036.
Market Trend: The NPU Inflection Point
Dedicated NPU silicon designed for industrial edge workloads is becoming a differentiating factor in vendor selection. Unlike general-purpose GPUs, industrial-grade NPUs offer lower power envelopes and optimized inferencing for vibration analysis, acoustic monitoring, and visual defect detection — all while generating less heat, a critical consideration inside sealed control cabinets adjacent to PLC racks.
Why the PLC Ecosystem Is Opening Up
For decades, PLCs operated as closed, hardened deterministic controllers — and for good reason. A production line stoppage costs thousands of dollars per minute. Yet the pressure to extract more data, predict failures, and optimize throughput without overhauling proven control architectures has created a receptive market for "sidecar" AI compute modules that sit adjacent to PLCs and ingest data over protocols like OPC UA, Modbus TCP, or EtherNet/IP.
ADLINK's position in this ecosystem is particularly strategic. The company already supplies industrial motherboards, data acquisition cards, and edge gateways deployed alongside PLCs from Siemens, Rockwell, Mitsubishi, and Omron. Adding Mobilint's NPU to that hardware portfolio creates a ready-made channel into brownfield factories that cannot afford a rip-and-replace control system upgrade.
Getac's Rugged Edge: Why Hardware Resilience Matters
One underappreciated variable in industrial edge AI deployment is environmental survivability. Factory floors subject electronics to temperature swings from -20°C to 60°C, high humidity, conductive dust, and continuous vibration. Getac's specialization in MIL-STD-810H and IP-rated rugged computing closes a critical reliability gap that many AI silicon startups overlook when designing for data-center conditions rather than shop-floor realities.
Industrial Edge AI: Key Market Data at a Glance
-
Industrial Edge Market (2025): USD 21.19 billion, projected to reach USD 44.73 billion by 2030 (CAGR 16.1%) — Source: MarketsandMarkets
-
Factory Floor Edge AI Industrial PCs (2026): USD 0.68 billion, forecast to hit USD 1.37 billion by 2036 (CAGR 7.3%) — Source: Fact.MR
-
Global Automation Market: Estimated at approximately USD 80 billion, growing 4–5% annually — Industry estimates
-
Edge AI Gateways for Machine Vision (2026): USD 690 million, reaching USD 2.82 billion by 2036 (CAGR 15.1%) — Source: Future Market Insights
FAQ: Edge AI and PLC Integration
Q: Does edge AI replace the PLC?
No. In nearly all current deployments, edge AI augments PLCs by handling non-deterministic tasks — anomaly detection, visual inspection, predictive analytics — while the PLC retains safety and real-time control functions. The two systems communicate over industrial protocols like OPC UA or Modbus TCP.
Q: What is an NPU, and why does it matter for industrial applications?
A neural processing unit (NPU) is a specialized chip optimized for AI inference workloads. Industrial-grade NPUs offer lower power consumption and heat output compared to GPUs, making them suitable for deployment inside control cabinets adjacent to PLC hardware where thermal management is constrained.
Q: Can legacy PLC systems support edge AI integration?
Yes. Most edge AI deployments interface with legacy PLCs through standard industrial communication protocols — no rip-and-replace is required. Edge AI compute modules function as "sidecar" devices that read PLC data, perform inference, and optionally write back insights or alerts.
The Road Ahead: From Partnerships to Production Deployments
The Mobilint–ADLINK–Getac collaboration remains in its formative stages, with integration engineering and joint solution blueprints expected to materialize over the coming quarters. However, the strategic logic is already validated by the market trajectory: industrial edge AI is not a speculative concept but a procurement reality, with factory operators increasingly mandating AI-readiness in their RFQ specifications for new control systems.
For system integrators and automation engineers, the message is clear: plan for architectures where NPU-driven inference modules sit alongside PLC racks as standard components — not optional extras. The partnerships being forged today will define which hardware platforms dominate that architecture tomorrow.
