Future of Robotics

WSJ published an article today on robotics and highlighted use cases in warehouses, meat packing, and other industrial settings. It's worth a read.

https://www.wsj.com/business/meet-the-robots-slicing-your-barbecue-ribs-338a7794?mod=hp_lead_pos7

Robotics has come a long way in a relatively short period of time considering the multi-level transformation required for adoption at scale. About 5 or 6 years ago, it was challenging for purpose-built robots to have a clear go-to-market pathway. It was also challenging to build robotic systems to handle complex tasks. While it continues to be challenging for robotics system to be adopted at scale due to factors such as organizational acceptance, pace of digital transformation, and budget availability, I am encouraged to see that we are moving in the right direction and that in certain areas, the adoption curve is flattening.

At the same time, we are still in the early phases of robotics (and AI which enables robotics). On the technical front, while robots are good at certain repetitive maneuvers needed in industrial settings, robots still need to learn more complex environments and become better at navigating them. One example I use is door opening. It is easy, right? It is easy for human because we have perception of depth, sound, light, environmental factors, and door configuration; our brains can instruct our hands to know how to open the door and navigate through. But doors can open toward or away from you. Doors come with different knobs and handles. Doors may be hard to find at night. Robots need to learn all of that.

What's encouraging is that the continued advancement of software, availability of real-world and simulation data, and more purpose-built models (reinforcement learning, multi-modal, vision, audio models) will accelerate the development of robotics systems. Robotics will move from "deep-tech or frontier tech" to "enterprise tech" with clear data, infrastructure, deployment stack and enabling capabilities for both industrial and everyday use cases. On the surface, we may be concerned about job security but the reality is that there are jobs that are better served and/or more safe if robots perform those tasks. Furthermore, as population declines with increasing percentage of an aging segment, we need robots to help us in the future.