Robots operating in unstructured environments are bound to run into situations that require human intervention. Having the right incident management system is key to meeting service level agreements (SLAs) to help prevent robots from failing constantly.
To say 2020 was a challenging year for all of us would be like saying water is wet. Duh. Instead, we turn to some of our favorite films for inspiration to explain why challenges are good for people and companies.
Gratitude comes in many forms, from the simplest “thank you” to the cashier at the grocery store, to a hand-written note or bouquet of flowers sent to an appreciated colleague on a job well done. In this crazy world of frenzied deadlines, headlines, appointments and endless video conference calls, it’s nice to have some time to sit back to reflect on where we’ve been and where we want to go.
The change of seasons and the beginning of the holidays for many offer a perfect opportunity to express our gratitude to those around us. While in previous years we may have skipped right past this time in our eagerness to get to the mall or begin our Cyber Monday shopping lists, 2020 and all of its insanity has reminded us of thinking about what’s truly important.
Today I’m happy to announce four new members have joined the InOrbit Board of Advisors, representing outstanding expertise in the field of robotics, cloud technology, supply chain management and venture funding. Our advisors help InOrbit accelerate our vision to help companies scale their robotic fleets through our cloud-based robot operations (RobOps) platform.
The four new advisors represent a wealth of experience across different fields:Martin Hitch, one of the co-founders of BossaNova Robotics, who helped the company pivot from consumer products to service robotics. During his tenure at Bossa Nova, he raised more than $40 million in venture investment.
Patricio Echague, co-founder and CTO at Split.io, is an entrepreneur and engineer passionate about data and high-performance systems. At Split Software, he’s changing the way software companies release features and add value to their customers. He was also part of the engineering founding team at RelateIQ, which was acquired by Salesforce.
Developing modern information technology solutions of any reasonable complexity will require that you integrate existing technologies at some point during the system creation process. Whether you’re just starting out or you’re looking to scale, you will face the decision of using an existing piece of software or building your own. The level of sophistication needed today can only be attained by building on top of other components, such as libraries, platforms or services. Just like in the old Western, experiences with these services can be good, bad, or ugly, depending on the providers you choose.
Even as InOrbit is building a platform to make it easier for robotics companies to focus on their own secret sauce, our software engineers have faced these decisions. In the course of our development, we have chosen services and components along the way. By choosing this path, we’ve also been able to add our own expertise, allowing us to augment those services to create a better platform for our customers.
Along the way, we’ve also had some mixed results that we’d like to share with you. We hope this lets you avoid some of the mistakes we made, as a way to help you speed up your own development process.
And should you choose to use the InOrbit platform yourself, we expect to be held to the highest standard.
The end of October brings out spooky and scary creatures of the night during Halloween festivities, but here at InOrbit we deal with ghosts, zombies and vampires almost every day. The different fantasy creatures give us a quick shorthand to the different artifact types we have available in our fleet simulation, the first of its kind.
In fact, these creatures are affectionately known as the “Hooli Robotics Horror Story Demo Fleet”, which operate as part of our fictional company, Hooli Robotics. Aside from the playful (scary?) names, there’s a deeper technical reason for all of this. There are many options, both open source and proprietary, for simulating an individual robot operating in a given environment. There are also solutions of creating variations, known as domain randomization, often used to accelerate reinforcement learning. However, what we found was missing was a way to simulate large fleets of robots, and to do that in a way that is both scalable and sustainable.
Most of our customers have enjoyed the Hooli fleet. While customers can’t directly interact with our spooky subjects (yet, but let us know if you want to meet them), in the spirit of the upcoming holiday, we'd like to give you a peek behind the curtain of our internal technology. Read on, if you dare, to learn more about the different types of spookiness that roams the hallways of Hooli…
Detecting insights from 3.8 million hours of robot monitoring data is elementary
For the last several years, we’ve all heard the quotes about data and information - it’s “the new oil”, it’s “the new science”, and that Big Data “holds all the answers.” Others have famously stated that “every company will eventually be in the data business,” or the real goal is to “turn data info information, and information into insight.”
But one of the best comments about data comes from our favorite fictional detective, Sherlock Holmes, who said (via Arthur Conan Doyle), “Never theorize before you have data. Invariably, you end up twisting facts to suit theories instead of theories to suit facts.”
As the world continues to battle the COVID-19 pandemic with varying degrees of success, the way we shop for groceries, clothes and other items has changed dramatically. Even as brick-and-mortar stores re-open for foot traffic, shoppers continue to buy items online, or shift their pickup to “buy online, pick up in store” and contact-free, curbside delivery.
For example, at a local Target near us, they’ve replaced the snack bar area with rows of shelves to store these pickup-in-store orders, but unfortunately the speed and efficiency of receiving your orders is not an improvement versus just walking through the store yourself and grabbing what you need on your “Target run.”
The problem lies in the method that physical stores are taking in filling orders - it’s the same that many distribution centers and warehouses employed when they didn’t have robots (or those warehouses that have yet to deploy them). A store receives an order, sending an associate out into the store to fill the order with a cart. The associate then returns to the front, where the items are placed on a shelf waiting for the customer to arrive.