For robotic applications in industry, automotive, and manufacturing, 5G technology enables high-speed, reliable, dense, and low-latency connectivity, suitable for both fixed-location and mobile use cases.
When discussing the exciting applications of 5G, healthcare, autonomous vehicles, drones, and consumer services requiring high bandwidth are among the most anticipated areas. However, 5G is a wireless technology that pioneers collaboration and high data transmission rates, featuring low latency and supporting the exchange of massive amounts of information, thus bringing disruptive changes to the machine and robotics industry.
There is a growing trend for robots to use artificial intelligence (AI) to guide their operations, and 5G's fast data transmission rate is conducive to enabling smooth communication between robots and AI servers, thereby enabling training or inference.
5G slicing
5G network slicing allows countless use cases with vastly different performance characteristics to share the same network infrastructure. Slices can be used for applications such as rescue robots requiring high speed and high bandwidth, autonomous vehicles requiring low latency, or collaborative robots in manufacturing requiring massive connectivity. Each application has its own dedicated network slice, enabling 5G technology to drive the development of innovative robotic applications.
One option for robot OEMs is to use 5G connectivity. These manufacturers can produce their own onboard chips (CoB) and assemble all the necessary components, such as modems, antennas, RF and IF front-ends, to engineer the interface. Another option is to use pre-built 5G modules, which are suitable for a variety of applications. For robot OEMs, configuring such modules in industrial robots is relatively straightforward, and they offer a viable option for adding value to robot systems through a range of features and functionalities.
Regardless of the choice made, these 5G connectivity features will need to be tested after being built into the robot. Even pre-tested modules may encounter coexistence, signal interference, or other wireless quality issues after being packaged with other components and integrated into the robot. These conditions can affect the overall functionality of the final product.
For OEMs, 5G testing is not a new requirement, but testing industrial robots can be quite challenging. One reason is that, to date, OEMs and their production lines have been configured only to manufacture and test robots without cellular capabilities. Therefore, integrating 5G testing requires updating testing equipment and fixtures.
5G technology offers flexible physical layer design, supporting a range of frequencies, bandwidths, subcarrier spacing, and modulation schemes. For many robot designs, 5G testing is still an emerging requirement, necessitating the addition of newer test cases to validate new functionalities within each testing domain. This work can be performed by R&D centers, design verification test centers, sampling inspection centers, production centers, or service centers.
All of this requires a re-examination of traditional testing methods and the use of cellular signaling testing or non-signaling testing techniques to ensure the quality of product and performance testing.
Signaling test and non-signaling test
The two cellular testing techniques (signaling testing and non-signaling testing) must be distinguished (Figure 1).
Figure 1: Simplified block diagram of signaling test and non-signaling test.
Although non-signaling testing is primarily geared towards and optimized for manufacturing applications, it is performed at virtually every stage of product development. This type of technology requires the use of RF signal generators and analyzers, focusing on the calibration and verification of the device's RF transmitter and receiver performance. This testing is performed in a processing mode that does not involve calls and utilizes chipset-specific test modes to measure pre-defined transmission patterns, thereby minimizing test time and reducing test costs.
On the other hand, signaling testing is widely used in product development phases such as R&D and design verification testing, and uses base station simulators to establish end-to-end user plane calls with the device under test (DUT). This approach has been extended to not only use a combination of control plane signaling and user plane traffic to measure the real-world performance of the DUT under realistic test conditions, but also to help verify each layer of the protocol stack to perform comprehensive device testing.
These two testing methods are designed to meet different testing needs and are optimized for use in different product lifecycles. However, the main basis for choosing between these two methods is parameters such as hardware design, testing focus, test type, test time, and test cost.
As the application scope of 5G continues to expand, the following section will focus on the importance of 5G testing in ensuring the quality of industrial robots.
quality assurance
When building products using custom-designed CoBs, it is essential to perform protocol and functional testing during the product design and development phases to ensure that the equipment undergoes extensive functional testing and software regression testing before being put into production.
However, test dynamics can change when building products using third-party cellular modules/RF subsystems/antenna modules. This is because even if the purchased modules are certified, there is no guarantee that the final product will function as intended in the field. The various components, once packaged together, can significantly influence each other and affect the overall performance of the product in a real network configuration.
Table 1 highlights some common questions.
Table 1: Some common problems and potential causes affecting 5G communication operations.
Each functional issue in the table significantly reduces communication reliability, thereby increasing latency and impacting speed. Given the interrelationships and non-exclusivity among the components, the importance and necessity of conducting 5G quality testing before and during production to comprehensively test the product are self-evident. Testing needs to verify:
1. Ensure basic device registration and call processes for end-to-end product functionality, as hardware components and system firmware can affect the robot's wireless performance;
2. Antenna performance is affected because defects in the final product's casing, implementation issues, and tuning errors can lead to signal loss and affect the antenna radiation pattern, causing communication failures between the robot and the 5G network.
3. Radio frequency performance ensures that signal quality does not degrade significantly under actual channel conditions and meets the acceptable signal transmission and reception levels specified by 3GPP and other certification bodies;
4. Data throughput, because under real-world conditions, the equipment must be able to handle various levels of user plane traffic and quality of service requirements;
5. Actual user experience, such as mobility and switching functionality;
6. Advanced features, such as embedded SIM cards and instant connectivity testing for multiple SIM cards.
Compared to traditional cellular technologies, 5G offers greater flexibility and richer functionality. Comprehensive product testing helps ensure overall product quality across different deployment models. Having outlined the importance of quality testing, the next equally important step is choosing a test suite that not only supports multiple functional tests but also improves the efficiency of parallel testing. Consensus between the production test team and test vendors helps vendors develop tools that support rapid Tx-Rx (transmitter and receiver) diagnostic testing and automated testing, thereby shortening time-to-market and controlling testing costs.
