Modest Tree is a data-centric digitalization company that specializes in leveraging advanced
technologies to transform the way companies collect, consolidate, and reuse training and operational datasets to propel their business goals. Working with original equipment manufacturers, airlines, and military clients, the company provides consultations, services, and digital tools to accelerate enterprise training and operational intelligence.
For research support with 3D object recognition, tracking, and digital mapping of real objects, Modest Tree partnered with Dr. Jiju Poovvancheri, a Professor in the Department of Mathematics and Computing Science, who conducts research on geometric deep learning on point clouds. Dr. Poovvancheri and his team of graduate students are currently working on a deep learning-based object recognition project that aims to accurately detect and identify industrial equipment, machinery, or automotive parts under varying lighting conditions and faded textures.
“Partnering with Dr. Poovvancheri has given us the ability to invite future innovators into the world of developing technology and allows us to share our experience and expertise while gaining valuable perspective from the next generation of computer scientists. Sharing knowledge between industry leaders and exceptional minds in academia is key to ensuring we stay on the leading edge of innovation to not only support our clients but to ensure the future generation has a strong foundation that they can continue to build on.“
Hamza Ayaz, Manager, Simulations and Training Solutions at Modest Tree
Dr. Poovvancheri’s research results will allow Modest Tree to incorporate machine learning algorithms into their training software and improve the accuracy of the trainees’ virtual experience.
This research was part of Modest Tree’s Tech Companion development. Tech Companion is interactive virtual support and electronic maintenance system that links VR tasks, training, record-keeping, and inventory management. The proprietary platform centralizes manuals, maintenance, and ongoing training while also supporting intelligent data management and analysis by leveraging advanced digitalization technologies into one comprehensive system, operational on multiple devices. Tech Companion starts with intelligently populating the software with existing maintenance instructions based on the client’s documentation. Using a data-driven framework, the platform inputs information to create intelligent forms and live document tracking. The proprietary platform centralizes manuals, maintenance, and ongoing training while supporting product warranty. With real-time, cross-platform manual updating, the software ensures industry knowledge transfer.
Tech Companion Modules Include:
Digital Workcards
Converts technical manuals into interactive work cards that provide procedural information on the task and collect data on the maintenance completed, streamlining service reporting and supporting compliance.
3D Training
Integrated 3D training lessons which the technician is able to access to support their maintenance task. Enables technicians to visualize the procedural steps for a task.
Remote Support
End-to-end encrypted chat, VoIP, and video functionality allow remote support between technicians across different device platforms. 3D annotation supports advanced service reporting.
Image Categorizer
Collects and categorizes images and components of an image taken during a service process. Images can be categorized based on asset, system, component, issue, etc. This provides a digital thread of an asset and is searchable for future compliance or analysis.
IoT Connectivity – Digital Twin
IoT Sensors provide remote sensor access and live data-to-machine reporting between your physical asset and your digital twin to enable live condition monitoring, data visualizations, analytics and remote diagnostics.
Parts Ordering
An application that allows technicians to pre-populate parts ordering information associated with a maintenance task, enabling procurement to receive timely information and see direct association to a specific maintenance task and individual.
Full OICE Report here
