Large scale manufacturing

Trimming and sanding on fiber-reinforced polymer (FRP) parts

Large-scale manufacturing is common practice in several industries: transportation, maritime, energy, aerospace and construction. Finishing operations, such as trimming, surface treatment, grinding, sanding and painting, are unavoidable for all types of materials to ensure components possess required geometrical tolerances and properties. For human operators, those tasks are tedious, leading to physical discomfort and the potential for repetitive strain injury. In this proposal, ”large-scale” is defined as a part that has at least one dimension equal to or larger than three meters. In this proposal, we will establish and demonstrate novel strategies for collaborative robotic systems performing two post-processing operations, trimming and sanding, on fiber-reinforced polymer (FRP) parts with at least one dimension equal to or greater than three meters, without part re-positioning.

Close Project

Autonomous collaborative sensing

For environment characterization and localization

We propose a framework to harvest the aggregated perception capabilities of several mobile robots for enabling efficient, dynamic, and real-time identification of critical surface features. In this context, a team of robots will coordinate and evenly distribute their efforts for mapping the environment, resulting in a real-time complete and dense 3D reconstruction of the part to be manipulated. The steps taken to achieve this are:

1) Real-time identification of critical surface features using RGB images and polarized images:

• Ding, Y., Ye, J., Barbalata, C., Oubre, J., Lemoine, C., Agostinho, J. and Palardy, G., 2021. Next-generation perception system for automated defects detection in composite laminates via polarized computational imaging. arXiv preprint arXiv:2108.10819 [accepted to CAMX 2021]

2) Sematic simultaneous localization and mapping

Close Project

Collaborative Planning And Control

For a mobile robotic manipulator team

To perform the finishing operations on large-scale parts, a team of robots will plan and execute motion and interaction tasks. In this project, we propose the development of a fully decentralized strategy based on rectangular map decomposition where the planning capabilities are informed by the control characteristics.

Close Project

xxxxx

xxxxxx

aaa

Close Project

Project Name

Lorem ipsum dolor sit amet consectetur.

Use this area to describe your project. Lorem ipsum dolor sit amet, consectetur adipisicing elit. Est blanditiis dolorem culpa incidunt minus dignissimos deserunt repellat aperiam quasi sunt officia expedita beatae cupiditate, maiores repudiandae, nostrum, reiciendis facere nemo!

• Date: January 2020
• Client: Southwest
• Category: Website Design

Close Project

Cognitive System For Task Completion

Task completion

In this work, we will leverage the avenues of robotic validation and human-operator validation for task completion. Specifically, we are interested in the robotic system to understand when the finishing task has been completed. To achieve this, we propose to develop a local-based evaluation methodology from information gathered from robot understanding and human experience. A combination of DNN algorithms and fiducial marker detection will be used locally at each agent to evaluate the completion of the task.

Close Project