Home Machine Learning Pavlo Pikulin, Founder & CEO of Deus Robotics – Interview Series
Machine Learning

Pavlo Pikulin, Founder & CEO of Deus Robotics – Interview Series

Share
Pavlo Pikulin, Founder & CEO of Deus Robotics – Interview Series
Share


Pavlo Pikulin is the founder and CEO of Deus Roboticswhich has developed an AI platform that connects and enhances the intelligence of warehouse robots from any manufacturer. The company also offers AI-powered robots that cover 90% of warehouse automation needs — and counting. With over 22 years of experience in AI and 6 years in robotics, Pavlo serves as an Expert on the Artificial Intelligence committee under Ukraine’s Ministry of Digital Transformation.

Can you share your journey in AI and robotics, and what inspired you to start Deus Robotics?

It all began with my dad, who’s always been a tech enthusiast. Even before I was born, he got me a personal computer — pretty rare in my hometown back in 1989.

When I was five, I wrote my first computer program. The idea of giving a task to a machine and watching it execute was absolutely fascinating to me.

At seven, I sold my first website for $20. Soon after, I started building and selling computers. By thirteen, I had written a program to bypass a captcha in a computer game, which later evolved into an automatic number-plate recognition system.

By fifteen, I was convinced that the future was all about robotics.

This realization set my path towards robotics and AI. However, at that age, I didn’t have the means to start my own robotics company, so I took a bit of a detour. At nineteen, I launched a game studio which grew into a global company with 60 million users worldwide. By 2021, we had 700 employees, and some of our games became the top downloads in their genre in the US.

But robotics was always on my mind and in my heart. I’ve always believed the world would be better if robots could handle all the tough, mundane, or dangerous tasks. By late 2017, I had saved enough to finally bootstrap my own robotics company and make that vision a reality.

It took me a year to do all the research in robotics I needed to start. After that, I gathered a team of incredibly talented engineers and programmers — some old friends, others new faces — and we launched Deus Robotics in early 2019. This past summer, Deus Robotics was shortlisted for Logistics UK’s Logistics Awards, Robotics & Automation Awards, and Supply Chain Excellence Awards USA.

What lessons did you carry over from your 15 years in the gaming industry to robotics and AI?

Working on games has given me valuable experience in many key areas, and it’s been rewarding to see how these skills apply to robotics.

Virtual Worlds for Robotics

Virtual worlds are a game-changer in robotics, offering a quicker and more efficient way to test ideas without waiting months for a hardware prototype. Instead of spending time and resources on physical prototypes, you can experiment in a virtual environment. Creating these virtual worlds is a lot like designing a game, where robots interact with their surroundings — this is where my game development experience really comes in handy.

Algorithm Transfer

When we were developing algorithms for robot path planning, as a team, we tapped into our experience from creating similar algorithms in games for character navigation. In our strategy games, hundreds of characters had to navigate through cities with roads and buildings — just like robots in our systems. This made transferring our path-planning expertise to robotics feel seamless and natural.

Complex Software and Analytics

Robots, much like games, are built on complex software. Developing a world-class game involves creating secure payment systems, social features like chat and groups, competitive elements, and a detailed virtual world with countless mechanics. One of our games had over 300 functional windows — more than you’d find in most banking apps! Games also handle massive loads, with millions or even billions of users performing countless actions, and this experience provides valuable insights for warehouse robotics.

Another crucial aspect is analytics. In games, we had to quickly and efficiently analyze user actions, managing high volumes of data. The same is true for robots, which continuously send vast amounts of data that need robust processing and analysis systems. Additionally, creating AI agents in games to engage players is similar to programming robots, as both involve crafting intelligent, responsive behavior.

In short, if you can build and manage a game development team, you’ve got what it takes to lead a robotics team as well.

How has your role in the Artificial Intelligence Committee under Ukraine’s Ministry of Digital Transformation influenced your work at Deus Robotics?

The Committee’s goal is to advise government authorities on AI, including robotics. I joined to provide expertise specifically in robotics and to consult on government requests. We’re also focused on education, regulations, and pilot AI projects in Ukraine. 

Through my work with the Committee, Deus Robotics found a way to expand our passion for advancing robotics in Ukraine beyond just warehouse automation. This allows us to make a broader impact.

Could you explain the unique AI brain developed by Deus Robotics and how it enhances the intelligence of warehouse robots?

Our robot control software, dubbed AI brain, includes key technologies:

1. On-Robot Systems:

  • A system that collects and analyzes data from all the robot’s sensors, helping it understand its surroundings, determine its location, and make decisions.
  • A control system that manages all robot mechanisms, such as manipulators, motors, lidars, and cameras, to execute the robot’s decisions.
  • A 3D recognition system that identifies obstacles using 3D cameras and a lidar recognition system that interprets what the LIDAR detects.

2. Centralized Management Platform:

  • Our AI platform connects all robots via Wi-Fi, coordinating their actions and enabling centralized communication.

In warehouses, our centralized system is crucial. It prevents collisions by tracking each robot’s exact location in real-time, allowing them to move quickly and safely.

This centralized system also integrates seamlessly with the warehouse management system, making it a key advantage and streamlining robot management.

Besides being connected, our robots are also getting smarter. Thanks to the adapters for different robots, our system can integrate robots from various manufacturers, enabling them to share information and coordinate tasks.

Previously, robots could only coordinate with other robots from the same manufacturer. Now, they can exchange data and work together across different brands.

What sets Deus Robotics apart from other companies in the warehouse automation industry?

Unlike other warehouse robotics companies, we’ve developed the heart of what makes robots truly effective: their brain. Our AI-powered brain connects with robots from any manufacturer, including our own, turning even the simplest robots into smarter, more capable machines. This kind of interoperability is something special in the industry.

This is especially important because, typically, robots from different manufacturers don’t play well together. Each robotics provider only covers a small slice of automation needs, leaving warehouses to choose from too many options. Each provider also requires costly and time-consuming integration, making it hard to switch if you want to try something new. Plus, when robots from different providers can’t connect or be centrally managed, it complicates operations and limits your ability to scale.

That’s where we come in. Our system makes it easy to bring as many robots as you need under one roof, all managed through a single, user-friendly platform.

With our solution, you get:

  • All automation needs covered under one system.

  • A single integration for all current and future providers, eliminating repeated integration costs and allowing you to switch providers as needed.

  • An easy-to-use management system for all current and future robots.

  • AI-enhanced, smarter robots that work together seamlessly (what’s important, our system can connect with humanoids, too).

  • Ongoing training and support for warehouse operators to ensure smooth operations.

And the best part? Our solution is truly all-in-one. We offer robots that tackle 90% of logistics tasks (and counting), paired with our AI platform that ties everything together. With our system, you can easily connect robots from any provider, simplify your operations, and scale your automation like never before. This is particularly important as only 20% of warehouses are automated.

How does Lidar technology enhance the navigation capabilities of your AGVs compared to traditional methods?

When our robot spots a floor tag, it navigates with pinpoint accuracy — within 1mm. But when the tag isn’t in view, accuracy can drop to around 20mm. That’s where LIDAR comes in, helping to keep navigation precise even between markers.

LIDAR adds another layer of information, which we combine with data from tags and gyroscopes. Each source gives us different insights with varying accuracy. Our specialized system merges all these inputs to pinpoint the robot’s position with high precision. The more accurate the data, the better the robot navigates, and precise positioning is key to smooth and effective robot movement.

Could you describe the role and benefits of your Robot as a Service (RaaS) and Software as a Service (SaaS) models for businesses?

Buying robots requires significant capital investment, leading to a payback period of one to three years. This also ties up the client’s operating funds and increases capital expenses.

RaaS (Robots as a Service) is a game-changer for businesses. It eliminates the need for capital investment, relying instead on operational expenses. The payback begins in the first month, thanks to the immediate boost in productivity. Instead of needing 2-5 times more people to achieve the same productivity, robots enhance the efficiency of existing staff—at a lower cost than hiring additional workers.

Plus, RaaS reduces risk. There’s no need to freeze investments, and expenses shift from CAPEX to OPEX.

What are the advantages of QR Code Navigation in AGVs, and how does it compare to Lidar?

Automated guided vehicles (AGVs) follow fixed routes, navigating via floor tags instead of LIDAR, and can’t deviate from their path. In contrast, our autonomous mobile robots (AMRs) use 3D LIDAR scans for flexible navigation without relying on floor markers.

In what ways do Deus Robotics’ solutions address the current challenges faced by warehouse logistics and the 3PL industry?

Currently, we offer a solution that transports goods from point A to point B, which is especially appealing to fulfillment centers. However, the need to move goods exists in any warehouse — whether it’s in manufacturing, healthcare, automotive, or retail. In the future, we plan to expand our product line to include robots that can handle tasks like packing, loading and unloading trucks, unpacking, and picking. We’re already working closely with robot manufacturers to integrate their products into our lineup.

Another challenge we’re addressing to make warehouses more efficient is providing a unified system for managing robots — a single integration, management, and enhancement point. Our clients don’t need to deal with multiple robot platforms. They only need to integrate with us once and can then use any robots they choose. While our current selection of robots is limited (we have 15 models to date), we’re constantly working to expand it.

What advancements do you foresee in the integration of AI and robotics in warehouse automation over the next 5-10 years?

One of the biggest challenges we face is figuring out the best way to automate each warehouse. Even though warehouses perform similar tasks, they differ in the details — like processes, staff, the types of goods they handle, and their goals. Some aim for maximum efficiency, others for cost-effectiveness. Planning the right automation solution and setting up the best business processes takes a lot of time and effort from experts. I’m convinced that AI will soon help us design warehouse automation in just seconds, making the process much smoother and more efficient.

Another exciting development is the rise of specialized robots designed for specific tasks. While the first robots were built for general use, as automation grows, we’ll see more robots tailored to unique needs. For example, inventory robots are just starting to appear since most warehouses only need one or two.

Humanoid robots are also on the horizon. These versatile robots will be able to handle tasks much like humans, thanks to advances in AI.

AI is also getting better at helping people with their tasks, like smarter voice assistants that can offer guidance. Plus, AI can optimize warehouse operations by analyzing data and improving efficiency. With access to digital warehouse data, AI can quickly identify and fix issues, making everything run more smoothly.

Looking at the bigger picture, AI will also improve how different warehouses work together, along with suppliers and logistics partners. This includes better scheduling, moving goods efficiently, and predicting what will be needed next.

All in all, AI and robotics are not just tools for automation — they’re the keys to unlocking a smarter, more connected future for warehouses and beyond.

Thank you for the great interview, readers who wish to learn more should visit Deus Robotics.



Source link

Share

Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Articles
Pavlo Pikulin, Founder & CEO of Deus Robotics – Interview Series
Machine Learning

3 Data-Proven Ways Companies Can Increase AI Adoption and Boost Productivity

As more companies explore how AI can drive productivity, one crucial aspect...

Pavlo Pikulin, Founder & CEO of Deus Robotics – Interview Series
Machine Learning

Microsoft AutoGen: Multi-Agent AI Workflows with Advanced Automation

Microsoft Research introduced AutoGen in September 2023 as an open-source Python framework...

Pavlo Pikulin, Founder & CEO of Deus Robotics – Interview Series
Machine Learning

Birago Jones, Co-Founder and CEO of Pienso – Interview Series

Birago Jones is the CEO and Co-Founder of Pienso, a no-code/low-code platform...

Real Identities Can Be Recovered From Synthetic Datasets
Machine Learning

Real Identities Can Be Recovered From Synthetic Datasets

If 2022 marked the moment when generative AI’s disruptive potential first captured...