Portable 5G AI Platform Revolutionizes Wildlife Conservation at Barcelona Tech Show
Barcelona, Sunday, 1 March 2026.
Technology company Canonical unveiled a groundbreaking portable wildlife conservation platform that fits in airline carry-on luggage yet powers comprehensive AI-driven monitoring systems. The airline carry-on sized system combines 5G networks, artificial intelligence, and cloud computing to connect drones, trail cameras, and research applications across remote conservation areas. Built on a 192-core Arm server with satellite connectivity, the platform processes wildlife video streams in real-time to identify animal species and track individual animals. This innovation addresses the critical challenge of monitoring vast wilderness areas where traditional infrastructure is unavailable, potentially transforming how conservationists protect endangered species.
Technical Architecture Powers Remote Conservation
The platform centers on the NextComputing NextServer AI 5G fly-away kit, featuring a 2U AmpereOne 192-core Arm-based server with up to 1TB DDR5 memory and 1PB NVMe storage capacity [1]. The system incorporates an NVIDIA RTX 4000 SFF Ada GPU and includes a 5G radio unit with external antennas capable of supporting up to four cells and 5,000 users simultaneously [1]. High-speed networking capabilities enable radio unit connectivity and internet backhaul through Starlink Mini satellite connectivity, ensuring global deployment flexibility even in the most remote locations [1]. This technical foundation enables the platform to process wildlife data locally while maintaining global connectivity for research collaboration.
AI-Powered Species Recognition and Tracking
The conservation platform runs open-source AI computer vision pipelines including YOLO, MegaDetector, and SpeciesNet algorithms on Ubuntu 24.04.3 arm64 LTS operating system [1]. These AI systems connect seamlessly with the TRAPPER conservation platform and BearID identification system, all operating atop Canonical Kubernetes infrastructure [1]. The system processes wildlife video streams in real-time to identify animal species and track known individual animals through a geospatial TRAPPER interface [1]. Visitors to the Mobile World Congress demonstration can observe live wildlife video streams being processed to identify animals and their species, showcasing the platform’s practical conservation applications [1].
Industry Collaboration Drives Conservation Innovation
The demonstration represents a collaborative effort involving multiple technology partners, with Canonical working alongside Arm, Ampere, NextComputing, and Software Radio Systems to develop the platform [1]. Academic and conservation partnerships include the Open Science Conservation Fund, the BearID project, and Universidad San Francisco de Quito, demonstrating the interdisciplinary approach required for effective wildlife conservation technology [1]. This collaborative model reflects broader industry trends, as evidenced by Arm’s continued support for conservation technology through partnerships with the WILDLABS Community since 2015 [2]. At the International Conservation Technology Conference held February 18-20, 2026, in Peru, Arm representatives presented sessions on “Running Conservation Technology Efficiently, from Edge to Cloud, on Arm,” sharing practical deployment lessons for AI sensors, camera traps, and bioacoustics systems [2].
Global Impact and Accessibility for Conservation Organizations
The platform’s cloud-native architecture and portable design address critical accessibility challenges facing conservation organizations worldwide, particularly those operating in remote environments with limited infrastructure [1]. The system’s design enables parallel AI inference and 5G workloads on the same hardware platform, significantly reducing power consumption and cooling requirements compared to traditional approaches [1]. This power efficiency proves essential for conservation teams working in remote locations where energy resources are limited [1]. The combination of portability, power efficiency, and comprehensive connectivity options positions the platform to democratize advanced wildlife monitoring capabilities for conservation organizations regardless of their location or technical infrastructure limitations [1].