Apr 14, 2020
Qualcomm products mentioned within this post are offered by Qualcomm Technologies, Inc. and/or its subsidiaries.
The use of Artificial Intelligence (AI) and the Internet of Things (IoT) have evolved rapidly over the last few years. In Deploying Retail on Mobile – Use Cases for Developers, we introduced you to how using AI and IoT could help your development projects personalize the retail experience. In this follow-up blog, we will look at how the fusion of 5G with AI and IoT, now coined “intelligent connectivity” or “hyper-connectivity”, are driving the digital transformation of retail. We’ll look at a few use cases to help you see the potential to take your development ideas to the next level of personalization. We’ll also look at how intelligent connectivity is even creating the opportunity for brand-new business models.
Why do we need intelligent connectivity?
To understand why the combination of these technologies is so significant, it helps to dip into the history of disruptions in retail. From providing the first catalogue that allowed customers to order from the comfort of home by telephone, to recognizing shoppers and their purchase history when they walk into a store using beacons, the goal has been to increase convenience and personalization. Currently, as shown in the diagram below, we are in the 4th Disruption of Retailing: Internet Retailing. The growth and impact of online shopping has affected the brick-and-mortar retail sector, leading them to actively investigate how to use new technologies to personalize their in-store experience in the hopes of winning shoppers back.
Overview of Disruptions in Retailing - Click to enlarge image
71% of shoppers agree with retailers that a personalized in-store experience is highly desirable, but the experience of shopping in a physical store needs to be a lot more compelling to beat the convenience of buying from the comfort of home on your laptop. AI and IoT developments have definitely gained ground, but the addition of 5G’s ability to operate at the latency, speed, and capacity to seamlessly support the next level of personalization technologies may be the 5th Disruption that retail is looking for. According to mobile communications industry group GSMA, “Intelligent connectivity marks the beginning of a new era defined by highly contextualized and personalized experiences, delivered as and when you want them.”
Intelligent connectivity use cases
In the three use cases below, you can see how intelligent connectivity technologies can help you develop solutions that create personalized experiences that benefit both shoppers and retailers:
Fitting rooms of the future
It's frustrating for shoppers to leave a fitting room just to get another size, or an additional item they didn’t realize they wanted. Smart mirrors that use intuitive technologies could act as personal digital assistants that provide shoppers with virtual access to all kinds of shopping resources to help reduce or prevent disruptions to the shopping experience. These interactive mirrors could use a combination of visual and gesture recognition technologies to identify the items being tried on, allow for the selection of alternatives, or request other selections to be brought in from the floor. Computer vision with deep learning could suggest styles that may be more flattering to the shopper’s shape, or change the lighting based on the intended occasion. Augmented reality (AR) could be added to superimpose the item in a different color or overlay virtual accessories. The mirrors could use embedded near field communication (NFC) to communicate with the shopper’s smartphone and act as a shopping basket that connects to the retail point-of-sale (POS) system for self-checkout, or even direct home delivery.
This type of intelligent, seamless experience requires streaming and real-time rendering from back-end systems. 5G’s ability to use a much wider spectrum at higher frequencies that can carry massive amounts of data at very high speeds with very little latency, or lag, makes it ideal for accommodating interactive AI/IoT applications. If you want to look at how to provide this type of high-quality personalized service, you can get started by reviewing the Qualcomm Snapdragon 845 Mobile Platform integrated with the Qualcomm AI Engine.
Navigating huge warehouse-style stores can be frustrating, and shoppers often have questions that product packaging lacks. Autonomous mobile robots (AMRs) can be summoned to a location to help shoppers quickly navigate to the merchandise they’re looking for. If the shopper has brought a product in, they can hold the product up to a camera to inquire about spare parts, or actually ask questions, and the AMR can act as a product expert using natural language processing (NLP) and deep learning. During their travels to and from shoppers, the AMR can simultaneously use computer vision to scan shelves to notify back-end inventory systems of misplaced items, or price tags that need to be changed. And since robots don’t need time off, they can also be utilized during off hours to carry out inventory counts using RFID sensors.
Latency is an enormous challenge for robots. In order for robots to function with intelligence levels that will seem natural and engaging to humans, 5G’s millisecond latencies can help with the analysis, reasoning, data fitting, and real-time optimizations needed for this level of AI. In addition, 5G’s ultra-reliable and low-latency service allows you to offload the AMR’s power-consuming AI functions to a real-time edge-cloud platform. This offers cost savings by lowering onboard computational and battery equipment drain on the AMR so they can run longer between charges. If you want to investigate these possibilities, try getting started with robotics. If you’re ready to start building these intelligent, power-efficient robots, have a look at the Qualcomm Robotics RB3 Platform.
Virtual reality merchandising
Determining product placement and an accompanying marketing strategy has traditionally been a very manual process involving repeated surveys and market research studies. The challenge is that it is generally a hypothetical situation for the shopper. Using virtual reality (VR) and eye-tracking technology, shoppers could walk through a virtual store and provide real-time consumer behavior insights. By being able to change configurations virtually, retailers can gather results more rapidly and affordably, and at larger scale, as shown in the collaboration project between QTI and Accenture.
However, VR apps can be interrupted by network performance, which affects the experience for the user. 5G can potentially offer a hundred times improvement in traffic capacity to a VR application. In order for applications like this to take hold and be successful, this increase will be critical. If you’re interested in starting a project that helps retailers solve similar problems, have a look at our getting started with XR (VR/AR) page.
New business model opportunities
The addition of 5G’s network data capacity, speed, and low latency to AI and IoT are also opening up new business models that could bring new opportunities for developers. Here are just a couple to get you thinking:
- Business-to-Home (B2H): VR and AR apps can be developed for showroom ordering with direct-from-warehouse home delivery.
- Robotics-as-a-Service: A fleet of robots, bundled with customizable software and hardware needed for executing specialized tasks autonomously, could be leased to retailers to support either front-end or back-end operations.
If you’re ready to look at developing a retail personalization project that utilizes the technologies of intelligent connectivity, a good place to start is by looking at our QDN connectivity options, and by downloading our QDN Connectivity Options for IoT Developers eBook. Remember that QDN is always here to help and we would love to hear what you’re working on, so contact us and let us know about your project!