The Foundational Dataset for Industrial Embodied AI
Forklift-Simulator is not just a training tool—it is the world’s leading standardized platform for capturing human skill acquisition in the high-stakes domain of logistics and material handling.
Next Generation of AI Partners
The future of automation relies on machines that can see, reason, and act with human-level intelligence in complex industrial environments. This Embodied AI requires one thing above all else: structured, real-world human behavioral data.
The Forklift-Simulator is not just a training tool—it is the world’s leading standardized platform for capturing human skill acquisition in the high-stakes domain of logistics and material handling.
We are actively seeking strategic partners to leverage this unique dataset to accelerate the Sim-to-Real bridge.
A Decade of Structured Human Skill Data
We have meticulously collected and standardized over 10 years of VR training sessions, providing an unparalleled resource for training, validating, and benchmarking embodied AI models.
- Multimodal Sensor Traces
Motion, gaze, controller inputs, decision timing, and physics telemetry.
- xAPI Skill Model
Structured, time-series learning curves that convert human performance into machine-interpretable reward signals for Imitation Learning.
- Physics-Rich Digital Twin
A proprietary, OEM-grade simulator environment for safe and accurate policy testing and Sim-to-Real transfer. - Ethically Sourced
All data is de-identified, anonymized, and compliant with global data standards (GDPR, KSA-PDPL).
“We have millions of hours of structured human performance data. It is ready for your model.”
Ready for the Open Ecosystem
We actively align our data and platform with the emerging industry standards to ensure seamless integration into the world’s leading AI frameworks.
- Meta Open-EV (Open Embodied Vision):
We are mapping our existing datasets to emerging Open-EV schemas, making our vision and action streams “plug-and-train” ready for Meta’s multimodal models. - NVIDIA Omniverse / Isaac Sim:
Our platform supports USD/ROS2 exports, enabling rapid, high-fidelity integration for synthetic data generation and robotics simulation within the Omniverse ecosystem. - Embodied AI Benchmarks:
We are working to establish the “Forklift-Skill Benchmark,” an open research task that provides a canonical, industrial challenge for labs worldwide.
Partner with Us to Accelerate Industrial Cognition
We are looking for partners across the AI landscape:
- Track learner progress, mastery, and safety benchmarks.
- FLS data supports required reporting for Perkins V and credentialing.
Let's build the future of embodied industrial intelligence together.
- Contact our AI Partnerships Team today to request a technical whitepaper and discuss data access.