This article was Part 3 in a four-part series on multi-machine agricultural systems, which addressed a specific challenge – how can connected machines be supported in limited connectivity areas?

  • Part 1 set the current landscape for multi-machine systems
  • Part 2 discussed impact of proper architecture and design
  • Part 4 addressed integration of fleet, task, and mission management

A significant challenge for multi-machine agricultural systems is that often they need to operate in remote areas that do not have reliable Internet access. For many applications, being dependent on a reliable Internet connection to operate is not an option. The last thing that farmers want to deal with is being shut down at critical times in the agricultural season because of connectivity issues. This means that scalable agricultural mobile machine systems must be designed to work in areas with no cell access or reliable Internet. This article describes the impact of this limited connectivity, and some of the strategies that JCA has used to address these challenges.

Internet Connectivity in Agriculture

Much of the land used for agricultural production is located in remote areas that are not highly populated. Infrastructure that enables modern world connectivity has been focused in urban areas, where most people are located, and then has slowly trickled out to rural areas. This, of course, is driven by the economics of this infrastructure: the more people it can serve, the more investment can be made in the infrastructure. IoT (Internet of Things) technology is changing the face of connectivity by extending the need for connectivity to machines and devices. As the benefits of increased agricultural production from coordinated machines is realized, the push for better connectivity infrastructure in rural areas is increasing.

There are many articles and efforts pushing for better connectivity in rural areas, as well as 5G connectivity, which aims to extend broadband connectivity worldwide. However, that is not the focus of this article. Rather, the focus here is to address how the benefits of connected agricultural machines can be realized with the connectivity available on farms today, while maintaining flexibility that can adapt to the technologies of tomorrow.

Farming operations are dynamic, unpredictable, and extremely time sensitive. A critical understanding in applying connectivity technology to ag machines is that downtime is very costly, so all efforts must be made to ensure that the main machine functions are not hampered by connectivity challenges.

For example, if the equivalent of a Microsoft Windows update hijacked the use of an agricultural machine needed at a critical time in the farming season, you can bet that Bill Gates would find himself on the business end of a forage harvester in no time. Finding solutions that can bring the benefits of multi-machine operation, without reduced reliability and increased downtime, is essential. This is challenging when much of the core technology has been developed for use in cases that are not mission critical; however, creative solutions can be found to leverage this technology and reap the benefits of applying it in a robust and reliable way.

Operator-Driven vs. Autonomous

The first question that must be asked when trying to enable connected agricultural machines is: What problem are you trying to solve? There is no one catchall for everything, so the specifics of the problem need to be understood, and the user-focused workflow approach discussed in Part 2 of this series can help to define an architecture for the system. Generally, however, it is useful to consider two broad types of systems that require different solutions to the connectivity problem: Operator-driven machines and autonomous machines.

Connectivity Strategies for Operator-Driven Machines

Operator-driven systems generally require connectivity for application data sharing, as well as support and service needs. Application data sharing helps the operator make decisions that optimize the machine operation. Support and service needs include occasional use connectivity for functions such as software deployment, log management, or real-time diagnostics.

Application data sharing typically makes the results of the operation of one machine available to a larger group of machines, which can be used to make decisions about optimal operation on the next task in the workflow. Sometimes, these task results need to be available in real-time for multi-machine systems that share a common mission, but in many cases for operator-driven machines, real-time connectivity is a nice-to-have, not a must-have. In most cases, machine systems can collect the operation data, and then provide it at a later time to make the next operation more efficient.

Applications that do not need real-time data sharing between machines can use a combination of logging and mobile devices to allow for seamless data transfer. A common strategy employed in many JCA-developed systems is the use of mobile devices (tablet/smartphones) as a means of managing the data transfer to allow for flexibility, which provides significant benefits.

Operating Multi Machine systems in limited connectivity areas

The mobile device acts as a user interface for the system, with a customized app tailored specifically to the machine and workflow. Operating data can be logged on the implement controller (often a JCA Falcon or Hummingbird controller), then transferred to the mobile device over a local wireless connection (Bluetooth or Wi-Fi). When a cell network is available in the field, the mobile device can be used to transfer data in real-time to a cloud system for multi-machine availability. When a cell network is not available, the mobile device can store the data locally, and then transfer the data to the cloud system at a later time where Internet connectivity is available (like back at the farmyard or home). This provides the benefit of using the mobile device’s data plan, which means a data plan fixed to the machine is not required. It allows for real-time connectivity through the mobile device when cell networks are available, but manages seamlessly situations where no cell connectivity is available. Software deployment to the system is also enabled through the app, which allows the user to choose the time and place for system updates (Bill Gates…take note).

However, some applications are better served with a direct connection from the machine to the cloud. This strategy is beneficial in situations where the mobile device does not fit with the machine workflow, or much of the data needs are not for the current machine user, but are for others in the multi-machine workflow.

This strategy obviously depends on available cell networks for multi-machine connectivity, but again using a strategy of logging data when networks are not available to delay transfers is possible, and is suitable for many systems where real-time data sharing is not critical.

Like what you’re reading so far? Subscribe to receive our Development of Multi-Machine Agriculture Systems content series directly to your inbox.

Connectivity Strategies for Autonomous Machines

Autonomous machines can provide additional challenges for multi-machine connectivity, largely because of the need for real-time connectivity and high availability. JCA has tackled these problems multiple ways with our autonomous framework (AFW) technologies. In some autonomous systems, multiple machines need to work together in a common mission, where activity is coordinated in a common work area.

In these cases, the JCA AFW architecture uses a multi-machine hub. The hub is responsible for machine coordination, dispatching, and overall mission management. Each individual machine has a subset of the mission assigned. Continuous communication is needed between the machines in the system, to HMIs (Human-Machine Interfaces) for user interaction, and for remote communication for mission planning data management, monitoring, and analysis. A simplified diagram of a multi-machine autonomous system is shown below.

A simplified diagram of a multi-machine autonomous system

Work areas for coordinated multi-machine missions can stretch over several kilometres for agricultural machine systems, and multiple types of information are needed to communicate locally within a work area. Some of this information is mission critical and needed by all machines, while other information may only be required from time to time, or between specific machines (or machine to HMI) within the work area. To span all of these needs, multiple communication types have been defined within the JCA AFW communication system, including:

  • Persistent – This is intended for long-range, lower data rate, and critical information. The information communicated over the persistent network includes critical health and status information, GNSS RTK corrections, and low-resolution mission progress updates. The persistent communication network operates within the 868 MHz-900 MHz frequency bands (depending on regional area) with a mesh network arrangement.
  • Proximity – This is intended for high data rate, but over a shorter range that may not span the entire work area. This is communication of high-resolution mission information, machine-to-machine close-range communication, and other high-bandwidth information (such as video feedback). The proximity network operates in both the 2.4 GHz and 5 GHz frequency bands.
  • Remote – Remote communication includes the cell network that communicates to the cloud system outside the work area. This is not required for operation, as it cannot be counted on in connectivity limited areas, but allows for remote monitoring, mission plan deployment, and analysis, when available.

A diagram showing the proximity and persistent range of communication in a work area is shown below.

A diagram showing the proximity and persistent range of communication in a work area

Each of the machines within the JCA AFW-based system, as well as the hub, runs a software stack that includes ROS (Robot Operating System). ROS facilitates various advanced functionality needed within an autonomous system. Communication within the proximity networks uses features of ROS2 that allow for multi-machine communication over a distributed network with robust security, authentication, and quality-of-service appropriate for mission critical communication systems using the DDS-security specification. DDS (Data Distribution Service) is a communication transport middleware that supports a publish-subscribe transport for messaging and is based on a defined standard with strong technical credibility, as it has been implemented in mission critical applications, such as space and flight systems, locomotive systems, and financial systems. A defined higher-level message protocol layer in the JCA AFW manages the combined proximity, persistent, and remote communication to provide seamless application development for the communication platform.

There are many more specifics regarding the multi-machine communication system within the JCA AFW that are beyond the scope of this article; however, the critical take-away is that this system has been shaped to the specific needs for robust, reliable, and secure communication for complex multi-machine systems in connectivity limited areas.

Adapting to the Future

In the development of multi-machine agricultural systems, JCA employs a strategy of using standard communication protocols and communication platforms wherever possible. Mature technologies and standards exist for much of the wired (on-machine) communication systems for agricultural machines, but there is still a high level of evolution in much of the technologies for connected multi-machine systems. In these evolving areas, JCA applies a strategy of leveraging technology that develops out of other industries where proven solutions and platforms have emerged, and then adapting this to agricultural systems in a way that can ensure flexibility as these technologies mature in the industry. While many of the specific challenges may be unique to agriculture, many other industries also face the need to adapt quickly as connectivity technologies mature. This common struggle to adapt has resulted in systems that are highly adaptable and a culture of integration through APIs (Application Programming Interfaces) that define interfaces between systems for increased compatibility. JCA has employed this same approach in our technology, allowing for technology and platform adaptability to be able to account for future connectivity improvements, and integration with third-party systems at many different levels.

Ready to talk?

Need help understanding how you can take advantage of these opportunities for your product? Talk to one of our experts and we can guide you through shifting from task management, fleet management and autonomous mission management into a multi machine system approach.

Next Part of the Series: Integration of Fleet, Task, and Mission Management

JCA has developed multi-machine systems across several different applications. Through this experience, we have faced many of the common challenges that arise with multi-machine systems and shaped our Autonomous Framework (AFW) technologies to address these challenges.

Read Part 4 - Integration Of Fleet, Task, And Mission Management

In Part 1 of the series, we described the current landscape and opportunities for multi-machine systems. Part 2 described multi-machine organization and architecture design, and now this part has provided an overview of multi-machine operation in limited connectivity areas. In the final part of this series, another specific challenge of multi-machine systems is addressed:

Part 4: Integration of Fleet, Task, and Mission management – Fleet and task management systems have developed along independent or loosely coupled paths. Multi-machine systems require both the integration of these and the addition of mission management, which considers the overall workflow of all machines in the system. A significant challenge with multi-machine systems is considering how this can be done effectively to result in an intuitive workflow for the user, while also leveraging existing systems, integrating with industry standards and common practices, and minimizing support infrastructure. This requires an understanding of cloud systems, IoT technologies, and the state and evolution of the agricultural industry for connected systems and is a major component of effective multi-machine systems.

About JCA Technologies
JCA Technologies was established in 2002 and has engineering and manufacturing facilities in Canada and the US. We partner with agricultural mobile machine OEMs to develop autonomous and highly-automated connected machines. We combine our technology building blocks with engineering and manufacturing capabilities to provide electrical, electronics, and software systems as part of customized mobile machine applications for innovative OEMs.