Reama AI Technology, a deep-tech research and deployment lab focused on developing reliable and responsible artificial intelligence solutions for Africa’s industrial sectors, has launched its LiveTwin pilot installation at a forestry plantation near Sabie, Mpumalanga.
The pilot, known as Reama LiveTwin, aims to advance intelligent monitoring of forestry assets through an orchestration of ground sensors, edge computing, and satellite data embeddings.
The pilot is designed to detect changes in tree canopy conditions, monitor vegetation encroachment near transmission lines, and provide early smoke detection capabilities that could support wildfire prevention efforts. At the core of the initiative is Reama AI’s vision to pilot one of South Africa’s first Living Digital Twins (LDTs), which is a continuously updating digital representation of the physical asset environment. The investigation aims to test LDTs as enablers of improved operations within large asset environments.
“Asset managers are under increasing pressure to make faster decisions across increasingly large and complex operations. Advances in AI, edge computing and earth observation are making it possible to generate real-time intelligence at a scale that was previously unattainable. LiveTwin represents an important step toward transforming how critical infrastructure and large assets are monitored and managed,” says Tebogo Mohlahlana, CEO and product architect of Reama AI.
“Our pilot at one of South Africa’s most significant forestry plantations integrates advanced technologies at the edge, where we are laying the foundation to make LDTs more accessible while fostering climate-resilient infrastructure, sustainable forestry, and AI innovation on the African continent.”
The research deployment leverages Intel Core Ultra 9 288V processors, Intel NPU 4.0, Intel Arc Graphics 140V and the Intel RealSense D456 Depth Camera. This hardware stack provides Reama AI with the tools to deliver powerful, real-time computer vision and audio classification capabilities directly at the edge.
Combined with Intel Edge AI technologies and OpenVINO software, the deployment enables low-latency analysis of forestry plantation environments without dependence on continuous cloud connectivity. The satellite intelligence stack leverages the latest Earth observation AI models, including Google DeepMind’s AlphaEarth Foundations, alongside complementary satellite technologies.