For warehouse robotics, orchestration software, and artificial intelligence to scale, cloud computing is key. GreyOrange Inc. today announced a collaboration with Google Cloud to develop GreyMatter DeepNav, which it said will dynamically manage and optimize autonomous robot operations at scale.
“We view the warehouse as a living ecosystem,” stated Akash Gupta, co-founder and CEO of GreyOrange. “GreyMatter doesn’t just connect hardware. It applies AI to orchestrate robots, people, and systems in real time.” “
“This new solution, trained on billions of real-world actions, will bring next-level intelligence to that layer — helping operators innovate without waiting months for payback,” he added.
Founded in 2012, GreyOrange said its GreyMatter and gStore continuously optimize automation, inventory, and workforce management for some of the world’s largest retailers and third-party logistics providers (3PLs). The Suwanee, Ga.-based company said its hardware-agnostic systems provide real-time visibility into omnichannel nodes.
GreyOrange said its software is available for diverse automation via the Certified Ranger Network and its Certified Partner Network of system integrators. It claimed that it can help customers reduce their cost per unit, eliminate lost inventory, ensure worker safety and productivity, and enhance in-store experiences.
GreyMatter DeepNav responds to need for innovation
Large warehouses use autonomous mobile robots (AMRs) to streamline and accelerate operations, from receiving to storage, picking, packing, and shipping. But despite the clear benefits of AMRs, many companies struggle with lengthy deployment times, noted GreyOrange.
Most warehouse robots are governed by hand-crafted rules, with minimal assistance from machine learning. Adding or adjusting AMRs requires expensive, time-consuming human intervention, which discourages innovation and prevents operators from fully adopting and scaling robotic systems, asserted the company.
“Together with Google Cloud, we’re not just accelerating robotics — we’re shaping the next generation of warehouse intelligence,” said Gupta. “Our GreyMatter warehouse orchestration platform optimizes up to 1 million AMR operations per minute, and with each action, we gain deeper insights into how robotic systems behave under real-world conditions.”
“For the first time, we’re capturing that learning in a way that OEMs and integrators can use directly — especially around how robots move and respond in real-world settings — the critical foundation for safe, scalable path planning,” he said. “Together with Google Cloud, we’re unlocking a new era of accelerated deployment and smarter warehouse automation.”
Google, GreyOrange collaborate on AI plus AMRs
GreyOrange and Google Cloud have been working together “for some time,” Gupta told Automated Warehouse. “We’ve been training different models to solve different problems, starting with smaller ones and moving to bigger ones. This has not been attempted before at commercial scale — it was only in research papers.”
Built on Google Cloud’s Vertex AI platform, GreyMatter DeepNav will enhance GreyOrange’s proprietary multi-agent orchestration capabilities — already proven in complex warehouse environments around the world, explained the partners. DeepNav will reduce the time and effort required to train new AMRs on intelligent path planning across fleets of robots from multiple vendors, they said.
“One critical problem statement that we have for GreyMatter is that the orchestration platform includes AMRs, QR-based navigation, and lidar,” Gupta said. “We’ve been working with Google to do this real-time navigation and orchestration in scaled heterogenous environments.”
With Google Cloud’s reinforcement learning capabilities, AMRs can achieve optimal navigation and task-execution paths within weeks rather than months, even in dynamic and high-density settings, said GreyOrange. Unlike traditional systems that struggle to scale beyond a few hundred units, GreyMatter DeepNav can handle robot operations larger the industry-standard ceiling of 300 units with ease and precision, the company claimed.
“It can adaptively handle all the paths for 1,000 AMRs in the field efficiently,” said Gupta. “When you introduce new layouts, robots, devices, process flows, or industries, we can just add them and train to new environments quickly. Every single day, every site will perform better. It can handle scenarios better because of hub-and-spoke teaching.”
GreyOrange listed the following features of GreyMatter DeepNav:
- AI-based path planning for real-time navigation in dynamic warehouse layouts
- Real-time decision making across heterogeneous agent types, including robot arms, PLC (programmable logic controller)-based systems, mobile manipulators, and automated storage and retrieval systems (ASRS) “We already support 15 devices from nine vendors,” said Gupta.
- Reinforcement learning–powered navigation and task assignment
- Support for dynamic task linking, interleaving, and service-level agreement (SLA)-sensitive execution
- Scalable control of thousands of agents in structured and unstructured environments.
Save the date for the 2026 Robotics SummitVertex AI to open applications, markets for automation
“With DeepNav, we can go into more complex CPG [consumer packaged goods] environments, pick cases and eaches, and have the capability to run a pallet warehouse with an e-commerce warehouse,” said Gupta. “Today, system integrators do condition analyses and simulations over a month; imagine this happening every single day in every single warehouse.”
GreyOrange is also working on bringing “conversational influence” into operations management, he said. Gupta added that the company is planning something “more complex than a chatbot — not just giving information but enabling authorized users to have a conversation and influence things happening in the warehouse.”
“The future of warehouse operations is intelligent, adaptive, and seamlessly orchestrated,” said Paula Natoli, global director of supply chain and logistics strategy and solutions at Google Cloud.
“With Google Cloud’s Vertex AI, we’re providing the foundational machine learning capabilities, including advanced reinforcement learning, that allow innovative partners like GreyOrange to develop solutions that can accelerate AMR deployment and scale to thousands of robots,” she said. “This represents a monumental leap in warehouse optimization, turning complex data into real-time operational excellence and unlocking unprecedented efficiency for global supply chains.”
GreyOrange said it plans to make GreyMatter DeepNav available in early 2026.
