Here are some key takeaways from Rand’s interview:
“While cutting-edge machine vision has been improving in leaps and bounds, other factors have contributed to the adoption of even simpler machine vision techniques.”
While there have been several interesting sensor improvements in the past decade, the most groundbreaking advancement has been along the software front through the advent of deep learning. At the same time, ongoing improvements in battery and motor technology have driven costs down allowing for the adoption of robotic systems that provide a very strong ROI to our warehouse operator customers. These robotic systems use inexpensive machine vision solutions with fiducial stickers, which robots can detect with ultra-precision.
“Our inVia picker robots use machine vision paired with fiducial markers so they can “see” and differentiate landmarks.”
Groundbreaking academic research helped fuel the popularity of LiDAR in the early 2000s, along with its use in autonomous vehicles in the DARPA Grand Challenge. While LiDAR can make obstacle avoidance and pathfinding easier, it can also create unnecessary complexity and expense. These sensor technologies work best in outdoor applications, where there are unknowns and inconsistencies in terrain. But LiDAR-directed robots get confused in indoor environments. They can’t differentiate between objects and robots get lost in the aisles of big warehouses.
2D machine vision in warehouse settings is cheaper, easier and more reliable than LiDAR. inVia Picker robots utilize a 2D machine vision with “fiducial” stickers. Our robots can detect the position and orientation of a fiducial sticker with ultra-precision. By sticking these fiducials all over a warehouse, robots can easily localize themselves.
“At inVia, we seek the most cost-effective technologies and components for our robots to make automation accessible to businesses of all sizes.”
Plain and simple, LiDAR technology is more expensive. Every robot that gets lost in warehouse aisles because of LiDAR re-mapping reduces customer ROI and can make the difference between fulfilling an order on time or a day late. Machine vision hardware is more affordable and reliable, providing a robust solution with less downtime and fewer errors.
“Warehouses that use robots to fulfill orders can supplement a scarce workforce and let their people manage the higher-order tasks like decision-making and problem-solving”
Machine vision helps automate rote tasks to increase labor productivity. Robots enable workers to take on higher-level tasks that require decisions and solving problems. Mobile robots can roam warehouses, picking, replenishing and managing inventory, undisrupted and accurately. Machine vision makes robotic systems affordable, driving greater adoption by small-and medium-sized businesses that used to be priced out of the market. They can now realize the samebenefits as their larger competitors, helping grow their business.
“When exploring robotic solutions, the most important step is to identify your biggest challenge and find the solution that solves it”.
E-commerce fulfillment is a very different animal than retail fulfillment, requiring its own specialized solution. inVia has developed a solution tailored to e-commerce fulfilment by creating proprietary algorithms to forge faster paths to retrieve more and different SKUs that come with e-commerce orders. E-commerce orders require different storage and retrieval plans, making it more labor-dependent, time-consuming and costly. Warehouses handling e-commerce orders want to reduce labor costs and order completion times for customers.
Source: Is LiDAR on its way out? The business case for saying goodbye