Toyota Material Handling Europe has announced the results of a physics AI collaboration (artificial intelligence applied to machines) with NVIDIA and SoftServe, demonstrating how AI-powered simulation and digital twin technologies can significantly accelerate, up to 50% the deployment of autonomous solutions.
The three companies have jointly developed a simulation-centered digital twin backbone for the development and validation of autonomous forklifts, driven by NVIDIA’s open Physics AI stack, which links engineering, simulation, digital twins, and materials handling operations. The collaboration combines Toyota Material Handling Europe’s track record in quality and lean thinking with the NVIDIA Omniverse libraries and the NVIDIA Isaac Sim open robotics-simulation environment, as well as SoftServe’s digital engineering capabilities, to validate autonomous-vehicle behavior in real-time physical simulations before client deployment.
“Through virtual commissioning and the digital-twin technology, we can guarantee quality and eliminate muda — waste — before deployment begins. By identifying and addressing issues earlier in the process, we can reduce implementation effort, accelerate value delivery for the customer, and offer automation solutions with greater confidence,” says Johan Rosén, Senior Vice President of Research & Development at Toyota Material Handling Europe.
Toyota Material Handling Europe has integrated the NVIDIA Omniverse and NVIDIA Isaac Sim libraries into its own workflows to develop new autonomous solutions tailored to each customer’s specific environment. Thanks to this simulation-first approach, Toyota Material Handling can optimize and validate solutions before deploying them at customer sites. SoftServe, an NVIDIA Elite Partner since 2022 and an NVIDIA Advanced Technology Partner since 2026, has designed the digital-twin environment that underpins this work, translating NVIDIA’s Physics AI stack into a simulation environment that Toyota Material Handling can use in production to ensure the zero muda objective: no waste of time or resources at the customer facilities.
“Zero muda sets a clear bar: eliminate waste before it happens, not after,” says Olha Moroz, EVP and Head of EMEA at SoftServe. “At SoftServe we’re proud to contribute to this through NVIDIA Omniverse and simulation-engineering developed for Toyota Material Handling Europe’s customers, enabling issues in the digital-twin world to be detected long before they can reach the forklift on the customer site.”
NVIDIA’s Physics AI Technologies to Improve Safety and Operational Performance
By integrating the Omniverse libraries, Toyota Material Handling will create digital twins to simulate plant layouts and validate the navigation of autonomous forklifts, safety logic, vehicle behavior, and client-specific edge cases. This integration enables autonomous vehicles to operate safely alongside people, manual forklifts, and mixed warehouse traffic.
The collaboration also extends AI into production through computer vision and intelligent agents based on NVIDIA technology for quality inspection, predictive maintenance, process optimization, and energy management. In this area, to train the camera-based pallet-detection module, synthetic data was generated with Omniverse libraries, augmented with NVIDIA Cosmos foundation-models’ open-world capabilities. As a result, the object-detection recall improved by more than 70% on real warehouse data sets evaluated.
The initiative aims to cut time-to-market by up to 40% and commissioning effort by up to 50%. In this way, it helps customers realize earlier value from autonomous material handling by minimizing on-site disruption during commissioning, improving solution quality before go-live, and shortening the path from design to a validated deployment. It also contributes to higher engineering productivity, manufacturing flexibility, and product quality, while reducing operating costs through AI-powered material handling at scale.
“Our ambition is to reach zero muda: eliminate waste that hampers productivity in material-handling operations,” says Ernesto Domínguez, President and CEO of Toyota Material Handling Europe. “By combining our lean mindset and automation expertise with NVIDIA’s Physics AI platform and SoftServe’s engineering capabilities, we can accelerate innovation, improve solution quality, and help customers achieve zero muda in their automated and autonomous operations.”
