During day 3 of TDK Ventures’ Digital Transformation Week (DX Week) conference, experts in the edge computing space focused on the technologies, hardware, and industry standards that will enable functionality required to make the greatest advances in the metaverse, healthcare, manufacturing, and more. Entrepreneurs, investors, and academics shared their expectations for the future of edge computing and the hurdles that must be overcome to get there.
Mike Vildibill, Qualcomm’s vice president of Cloud Edge AI, invoked Elon Musk as a catalyst for innovation and innovation and for “identifying a grand challenge of autonomous driving as something that he has put tremendous resources behind.” Musk’s work at Tesla provides “excellent examples of application pull that have an impact on the speed of innovation to support that grand challenge.”
To truly speed society toward universal edge computing solutions, industry and research must move beyond use-case-based projects, Vildibill said. He urged entrepreneurs, investors, and engineers to accept the grand challenge of “pulling the industry together in those directions. Industry standards or interoperability or the ability for multiple technology innovators to mix and match their technologies together to yield or provide an end product,” he explained. “A really difficult challenge between the initial vision and the [product] demonstration — it’s usually vertically integrated — is moving it toward something that an entire industry can go after and make real, compelling, global-scale impact. That takes a lot of energy and a lot of cooperation.”
A similar approach may be necessary to realize the metaverse’s full potential, Vildibill said.
“Our imaginations are being sparked; we’re seeing new possibilities,” he explained. “The art of the possible is changed with meta, and it’s going to be interesting to drive those use cases and technologies but also to make sure that they are not just vertically integrated point solutions but rather are technologies that industry as a whole can come together around and drive a very broad, impactful set of deployments.”
Innovators must expand their mission beyond “just doing the proof or the single demonstration, but the hard work of making that something that industry and society as a whole can drive forward.”
Much of that forward-thinking will be based on the exploitation of 5G wireless computability on the edge. But at least one panelist was already looking ahead to the next generation with even lower latency, blazing speed, massive throughput, greater reliability, and more seamless availability and user experiences. Dean Brenner, chairman of the Federal Communications Commission’s Technological Advisory Council, acknowledge that 5G still has a bright future with continually enhanced standards.
“In the past, a G has had a lifespan of about 10 years, and the first 5G deployments occurred in about 2019. So, if that timeline were to hold, over the next 10 years 6G will be conceptualized, planned, designed, and in an early stage of deployment.”
No one can make a foolproof guess about what 6G will look like or be able to accomplish, “but we know every single country in the world is going to be the leader in 6G.”
He said the race for current-generation supremacy commenced “toward the ‘end of the beginning’ of 5G. With 6G, even now, the race is on.
He said it is also reasonable to assume that 6G will bring exponential steps forward.
“For each G, there’s been a more than 10X improvement in capacity,” he said. “For the billions of dollars of investment to occur in a new G, it’s got to be a LOT better than the prior G. So, we know 6G is going to have to be disruptive in providing the kind of connectivity we can only dream about today.”
And, with so many new participants in the space and so many industries seeking to exploit and benefit from it, cooperation will be key to prevent fragmentation and chaos from reigning.
The public sector, of course, includes the FCC and other global, national and regional regulators.
“The way a new G gets planned — in the private sector you have individual companies, groups of companies operating through standards bodies, and you have universal adoption that takes giant capital investment, a lot of time, and a lot of bets by venture capital concerns. The automotive industry; industrial; [internet of things]; those are all relatively new players, and those companies that are in those various verticals are not going to take a vacation from 6G,” Brenner said.”
As an entrepreneur, EdgeQ Founder and CEO Vinay Ravuri, said whether 5G- or 6G-based, he views technology transformations in two phases: development and application.
“For the cloud to get to where we know it today, first somebody had to develop and deploy it, and services could get built upon it as application of the cloud,” he said. “In the same way, we are in the technology development phase of 5G. Even the deployment of a network is still part of this phase.”
Ravuri said 5G will not begin to be utilized appropriately until businesses begin offering services that use it. Around the same time those use cases hit their stride — five to 10 years from now — 6G developers will “compile requirements and play with different fundamental technologies like terahertz processing, computer interface, and neural links. Then comes the actual technology development so it becomes something greater than a research project.”
Still, 6G’s possibilities allow us to dream, Ravuri said. It could prove an enabler for a variety of Jetsons-like labor-saving devices. “Not just watches, but things that deliver medicine on their own into your system. And human-machine interfaces — “quadriplegics would be the first beneficiaries, but we’re lazy and would love to just sit there and let the computer do the work.”
Useful Sensors CEO Pete Warden said miniaturization can facilitate edge computing, 6G development, and a refocusing of efforts onto interoperability.
“Compute keeps getting smaller and smaller and more energy efficient,” he expanded. “Continuing that trend over the next 10 years, I’m hoping we end up with devices that are smaller than a grain of rice, getting down to ‘smart dust’ that includes enough battery power to run things like machine learning computing.”
He envisioned scattering tiny devices in fields where they would monitor crop conditions and prescribe herbicides, pesticides, and fertilizers to maximize crop health and yields. These microcomputers could inform more ecologically friendly agriculture.
He said these components could be mixed into concrete to monitor stress on buildings and bridges for scheduling maintenance, repair and replacement.
Even inexpensive peel-and-stick sensors could employ machine vision technology to detect leaks and other defects in manufacturing equipment over the long term.
Elana Lian, an investment director with Intel Capital, said miniaturization offers an investment opportunity.
“There’s a different level of automation as we approach the smart dust level,” she said. “Eventually it will enable all five levels of automation, starting at level 1 — a single automated device, a machine that does not have the capability to sense and adapt to changes in the environment. Moving to more powerful machine learning algorithms and hardware, level two is partial automation. The machine can sense the environment and make minor adjustments. Level 3 is more conditional automation. The machine can control and monitor all the environment but still needs humans’ attention and interaction. Level 4 will be more high-automation with full autonomy in a ‘defined area’ or specific solution. Once we get to level 5, we will have full autonomy in all situations. That’s linking our minds and using our neuroscience to control the light switch and thermostat.”
The concept of true autonomy intrigued the panelists, and that will be the focus of our third article focusing on DX Week’s Edge Computing panel.