How to combat food insecurity with flexible deployment of 5G, AI and edge solutions in agriculture

Commercial context

Food insecurity remains a global challenge of the first order, affecting 30% of the world’s population (2.4 billion people) in 2022 – with 11% being classed as severely insecure. Across the global agricultural sector however, unsustainable practices are actually reducing production, with deficiencies in resilience to climate change, inflation, labor shortages and supply chain disruption.

Climate change presents especially pressing challenges to farmers must adapt quickly - with newly erratic weather patterns, higher temperatures, greater demands on disease and pest control systems, and the urgent need to introduce resilient new crop strains. Farmers must monitor increasingly complex and diverse environments with smaller qualified workforces, and require real-time data during growing seasons to monitor soil conditions, crop cycles and animal health – yet many have limited access to cellular networks due to their inherently rural locations.

Connected agricultural technology (agtech) has hitherto been limited – but that picture is finally changing, with Global AGTech Initiative noting that funding of agtech in 2022 rose 15% YoY to $7.5bn. Yet much existing infrastructure is still focused on straightforward data analytics, with farmers required to manually input data and rely on low-cost sensors, and LPWA connectivity rather than 5G. Modern farming technology – such as AI solutions, drones, and robots – requires the extreme data throughput, massive connectivity, ultra-low latency and high reliability that 5G provides.

Now then is the time to turn to 5G, edge computing and AI to revolutionize agtech. With 5G connectivity farmers can draw on a range of compute options to enhance productivity – for instance drones and AI robots to enable precision farming and automation, digital agronomists and veterinarians, and remote monitoring of crop health, animal behaviour and irrigation.

The solution

This is where the Intelligent edge for sustainable agriculture - Phase II Catalyst comes in. The project is refining a scalable blueprint that is open, to allow any application or service provider to join, flexible enough to deploy on cloud and remote locations, low-latency for high processing needs, and portable to any cloud or edge provider platform. It’s also operationally consistent to simplify scaling and knowledge-sharing, secure, and easy to deploy.

This amounts to a unified platform for agricultural applications relying on high-speed connectivity – supporting multiple end devices, diverse SLAs, and various QoS requirements, using TM Forum standards – enabling farmers to select and customize robotic and drone services easily. Farmers can draw on robotic or drone services, satellites, high-quality dynamic visuals and computer vision, and advanced machine learning capabilities for real-time data insights and inferences. 5G edge-enabled AI decision-making is made in a closed-loop environment with agricultural robots for the defined use cases, with applied dynamic integration of orchestration and 5G network slicing, while defining different service classes with appropriate quality of service levels.

This end-to-end blueprint integrates the use of TM Forum Open APIs, open standards and open-source technologies to design a dynamic marketplace covering multi-faceted business models for smart farming covered by hybrid networks. Now in its second phase, the project plans next to incorporate digital twins to enable smart farms to be composed and optimized virtually, prior to physical implementation, eliminating design inefficiencies while addressing sustainability goals. The predictive analysis made possible by this which will transform the way farming is done – helping to conserve natural resources, while meeting the global challenge posed by food insecurity.

Applications and wider value

This year’s iteration of the Catalyst has shown a proof-of-concept for the blueprint’s application in smart weeding, crop health monitoring, smart irrigation and animal health monitoring, with demonstrated impact on UN SDG metrics such as water savings, energy consumption and carbon savings. The results are striking – the smart weeding use case for instance shows a reduction of up to 20% in chemicals used, while smart irrigation achieves up to 30% less water use.

The driving purpose behind the project is to make is easy for farmers to onboard applications and services for multiple use cases, enabling not only more productive and sustainable practices in that instance but building a connected value chain in agriculture – between farmers, agronomists, manufacturers, retailers and distributors – which is present in other sectors but still notably missing in this one. The effect will be a considerable step towards net zero even while food production increases, as do CSP revenues with expansion into the agricultural vertical.

Vimal Mehta, Senior Technology Architect at TELUS, explains that “to have a meaningful impact at the farm level, solution providers must deliver AI, IoT, computer vision, robots, and low-latency connectivity services at scale to increase adoption. This requires a paradigm shift in how we trade and deliver solutions. Our Catalyst leverages the capabilities of its participants and the TM Forum ODA framework and Open APIs, and recommends new industry guides and specifications to the TM Forum – the resulting blueprint enables multiple vendors and operators to work together and deploy complicated solutions seamlessly and transparently to the farming industry.”