Use Cases
Demonstrations are already targeted to show the benefits and maximize the impact of TIMING. The demonstrations are planned to be carried out in the TIMING experimental environment.
End-to-end architecture to demonstrate the TIMING solution.
For experimental demonstration and KPI validation of the TIMING solutions, the architecture depicted in the next figure will be deployed. The architecture consists of two hybrid TSN domains with Wi-Fi and Ethernet TSN nodes, which are interconnected by a no-TSN network, representing an operator metro network. The aim is to carry out experiments and demonstrations for intra-domain TSN and inter-domain TSN scenarios.
Just for demonstration of the TIMING solutions, industrial applications will be showcased. Specifically, two use cases have been conceived.
USE CASE 1: Virtualized PLC for line follower AGVs
Unmanned transport vehicles, also known as Automatic Guided Vehicles (AGVs), are utilized in the industrial sector to replace manned industrial trucks and conveyors. Each AGV is controlled by a programmable logic controller (PLC), which is responsible for managing the internal control loops, which includes gathering data from the guiding sensors and making suitable control decisions. The PLC is now co-located with the AGV due to the tight latency requirements, therefore all communications with the sensors and engine are wired.
One of the most used AGV types in industry is the line follower, which is normally equipped with a sensor that provides the relative position to a path established on the floor. The sensor indicates if the AGV is either at the left or at the right of the path and the separation distance. This way the controller can take the proper control actions to follow the line. The next figure shows an example of line follower AGV.
The application of TSN to ensure low latency communications could allow the PLC control logic to be moved out of the AGV, centralizing AGV control. The approach entails splitting the PLC into an on-board Slave PLC (sPLC), which collects data from sensors and physical inputs and connects to the motors, and a virtual Master PLC (mPLC) running on a server, which is in charge of processing all data, making appropriate control decisions, and sending them back to the sPLC, which translates them into signals to command the motors. Specifically, the sPLC serves merely as a physical signals gateway, whereas the mPLC virtualizes and executes all control decisions. This control centralization enables more intelligent decisions to be made, as well as a more adaptable and changeable factory, which is one of the benefits of Industry 4.0. It is also predicted to give considerable benefits in terms of redundancy, cost savings, scalability, lower power consumption, and hardware independence as a result of mPLC virtualization.
Easybot AGV (by ASTI Mobile Robotics)
To make this transition possible, the TSN technology must enable proper synchronized communication between the sPLC and the mPLC, ensuring that sensor data and control actions are delivered on time and reliably. The goal of this use case is to first establish TSN's capacity to give appropriate performance, and then to experiment with a wide range of scenarios to determine under which situations the AGVs will behave as expected. In this use the mPLC will be connected to the sPLC thanks to TSN.
For this demo, the sPLC will be equipped with a Wi-Fi TSN modem and connected to a Wi-Fi TSN AP, whereas the mPLC will be connected directly to an Ethernet TSN switch.
USE CASE 2: Virtualized navigation stack
In general, AGVs are equipped with guiding and localization systems. The guiding systems allows to know the deviation between the AGV and the path defined in the working environment. There are different types of guiding sensors: optical, magnetic, etc. The localization system detects points of reference placed in the working environment, with a known specific location on the plane. These reference points may be sensed by an optical, magnetic or electromagnetic device. This way the AGV can know its own location. The accuracy of the localization system is usually determined as half the distance between two points.
The guiding and localization functionalities of the AGV can be implemented by only one device, such as a navigation system based on laser reflectors or the SLAM navigation systems. However, this solution increases the cost of the AGV. The navigation stack is one of the most computationally demanding layers of AGV control. Virtualization of this layer and execution as a multi-access edge computing (MEC) application would allow sharing computational resources among several AGVs, reducing the unit cost of the equipment and the design of more advanced navigation strategies that exploit the fusion of information collected by different AGVs.
EBOT AGV (by ASTI Mobile Robotics)
To test this approach, an AGV similar to that shown in the figure will be used. This AGV is equipped with a navigation system which combines QR-codes and odometry information, and provides an excellent maneuverability. Thanks to its compact design and omnidirectional and turn on spot technology, this AGV can perform longitudinal and transversal movements, enabling greater agility and efficiency.