US-based robotic security services provider Asylon has announced a collaboration with NVIDIA to advance physical artificial intelligence (AI) in autonomous security systems. The partnership introduces DroneIQ Overwatch, a new AI-powered capability within Asylon’s DroneIQ platform, designed to enhance real-world security operations where humans, robots, and AI systems operate in coordination.
Scheduled for rollout in 2026, DroneIQ Overwatch adds an intelligent analytics layer to Asylon’s existing robotic security ecosystem. Currently, the company’s systems are managed through DroneIQ and supported by its 24/7 Robotic Security Operations Center (RSOC). With the addition of Overwatch, the platform will continuously analyze live video feeds and operational data, automatically identifying anomalies and flagging potential security events for human review.
The solution aims to address limitations in traditional monitoring environments, which often rely heavily on manual observation or fragmented alert systems. By acting as an initial layer of situational awareness, DroneIQ Overwatch is designed to accelerate threat detection while maintaining human oversight for validation and response in critical situations.
Damon Henry, CEO of Asylon, highlighted that the evolution of security lies in augmenting human capabilities with intelligent systems rather than replacing them. He described DroneIQ Overwatch as a step toward “Physical AI,” where autonomous machines function in real-world environments with humans remaining integral to decision-making and accountability.
The platform is powered by a hybrid edge-to-cloud AI architecture built on NVIDIA technologies. Asylon’s robotic systems incorporate NVIDIA Jetson modules for on-device processing, enabling real-time inference directly at the edge. Complementing this, cloud-based GPU infrastructure provides scalable analytics, centralized model management, and continuous AI refinement.
This integrated architecture enables real-time anomaly detection at the robotic edge, scalable analytics across fleets, and customizable detection models tailored to specific operational requirements. It also supports continuous improvement through centralized updates, ensuring adaptability to evolving security challenges.
The collaboration reflects a broader shift in the security industry toward embedding intelligence within physical systems. As enterprises face increasing operational complexity, workforce limitations, and rising costs, the convergence of robotics and AI is emerging as a key enabler of scalable and efficient security solutions.
