As automation in manufacturing and logistics operations modernize, mobile equipment is increasingly being used to aid in material logistics and in the assembly of finished goods. Automated Guided Vehicles (AGVs) can be seen carrying pallets of raw goods to manufacturing cells and then delivering finished goods to inventory or shipping. In modern automotive assembly plants, autonomous vehicles are often outfitted with lifts to position vehicle chassis for easy access, so assemblers can perform their tasks ergonomically. Additionally, mobile machinery now allows manufacturing to be scaled effectively in a smaller footprint, helping to drive long-term costs down.
One of the key challenges in integrating mobile equipment is making the critical data within these systems available so they can be effectively monitored and controlled within the environments they operate. Even the most compact AGV can be a blend of different network technologies so it will be beneficial to understand these technologies so strategies towards making data available can be developed.
To understand how networks come into play within AGVs, it is good to know the basics of how they work. While not covering every detail of the control system, we will discuss the main aspects.
Like many automated systems, there is a controller that serves as the main intelligence of the vehicle. These are commonly based on microcomputers or programmable automation controllers, and are responsible for communicating with the Warehouse Management System (WMS) or Traffic Management software. Through this communication they can receive tasks such as to deliver some raw material to a machine and communicate back when these tasks are completed. The controller also interacts with the peripheral functions in the vehicle that control steering, drive, power and collision avoidance.
As the brains of the vehicle, the controller needs to be able to support all communications technologies implemented within the vehicle in addition to those supported at the WMS level. These are very often different as the needs of information-based applications are not the same as those needed for control.
Within an AGV, the drive and steering controls manage the prime movers of the vehicle. Servo drives and motors in combination with feedback sensors are used for side, forward and reverse movements of the vehicle and these movements need to be balanced by the control system. The drive and steering controls within a vehicle are often integrated using CAN or CANopen as a control network. This has some basis in history as CAN has been heavily used in commercial and passenger vehicles for decades, so its performance and reliability are well understood. Additionally, CAN is known for its relatively low power consumption versus other network technologies, so it is well suited for use in battery-powered systems such as AGVs. It is also not unusual to see Ethernet-based control networks inside the AGV as well.
Another critical control system within the vehicle is power management. This would include the batteries themselves and the Battery Management System (BMS), which manages feedback from multiple sensors to monitor the battery state and protects the battery from operating in an unsafe manner. The BMS, like the steering and drive controls, is quite often supporting CAN as its control network, again for its historical use in vehicles and low energy consumption.
Lastly is the collision avoidance system which is responsible for detecting the presence of people or objects in its path then prescribing route or speed adjustments. This typically consists of multiple safety-rated proximity sensors or vision systems designed to give the vehicle eyes into its environment and protect both people and other equipment. Many of these devices are still using safety-compliant digital I/O as its primary interface but are increasingly being used on networks supporting safety protocols such as PROFIsafe, CIP Safety, FSoE and others.
In addition to handing communications with the internal control networks, the AGV controller needs to be enabled with connectivity to external systems. This is achieved using wireless technologies with 802.11 Wi-Fi, typically 802.11n, being the most common interface used for connectivity. Additionally, AGV controllers will often need to be enabled with 802.11r alongside 802.11n to allow for near-seamless roaming between access points. The need for low-latency real-time control and IoT connectivity may also see the emergence of 5G in AGV controllers.
Designing AGV control systems that accounts for all forms of connectivity is one of the most important challenges that developers and integrators face.
Approaches towards developing connectivity for AGV controllers differs based on the platform being used. For starters, let’s take CAN into consideration for steering, drive and power controls.
With microcomputer-based or embedded AGV controllers, CAN connectivity is typically developed making use of the peripheral interfaces supported by the microcontroller on the main board. Common interfaces include synchronous serial interfaces (SPI, I2C, etc.), USB and in the event of PC-based systems PCI-e Mini. For embedded developers, adding connectivity from ground up can be a slow-moving process as this requires the implementation of CAN transceivers, supporting electronics and communication protocol stacks. Development, testing and certification often takes several years to complete. Modular solutions with built-in CAN intelligence are available and can significantly shorten time to market.
As mentioned earlier, some AGV control systems center around off-the-shelf automation controllers provided by companies such as Rockwell, Siemens, Beckhoff, Schneider and others. The challenge in adding connectivity for the steering, drive and power controls is that CAN connectivity is not native to these platforms. This can be accomplished using protocol gateways that support both CAN and the controller’s native network protocol, or in the case of some automation controllers, an embedded solution that fits into the backplane of the controller.
For wireless communications, its heavy commercial and home use means that there are numerous embedded solutions for developers, both chip-based and module based. In terms of off-the-shelf solutions for interfacing to automation controllers, numerous Ethernet to wireless options exist but it is critical that these solutions be ruggedized for industrial applications and provide the throughput needed for critical data. Applications where data is transmitted wirelessly from safety controllers require extremely reliable interfaces.
Lastly, the need for greater data transparency means that AGV controllers will need to support connectivity to information-centric applications. For embedded controller developers, it can be challenging to implement standard information protocols (ex. OPC UA, MQTT) securely and shape the data in such a way as to be properly understood by these applications. Modular solutions for embedded controllers with support for these protocols can mitigate some of these challenges. For systems using automation controllers, edge gateways can be used to accumulate and add context to data without taxing the control system..
There are a lot of considerations when it comes to addressing connectivity on mobile equipment but fortunately solutions exist to ensure that critical data will be available when needed.