Subject to the finalization of all contractual details and requirements, the 8 selected Technology Projects from the ARM Fall 2017 Project Call are briefly outlined below.
Vision-Based Cleaning of Complex Structures with a Lightweight Compliant Arm
Principal Investigator: National Robotics Engineering Center (Carnegie Mellon University)
Project Description: This project centers on the development of a robotic system for surface treatment in a confined space. Once fully developed, ARM members will be able to leverage the software modules to enhance existing robotic systems with advanced manipulation capabilities. The modularity of the delivered software will enable future derivatives such as standalone visual inspection, generic surface treatments, advanced surface treatment, 3D mapping in confined spaces, and more.
Collaborative Robotics to Foster Innovation in Seafood Handling (FISH)
Principal Investigator: Northeastern University
Project Description: The FISH project will advance a robotic system that can reliably grasp, place, flip, and maneuver seafood. The project centers on the development of perception and planning algorithms that will identify and characterize different types of seafood and grasp them in constrained, collaborative environments alongside human workers. Perception in this project is critical due both to the need for human collaboration in the environment and due to the unpredictability of the objects. The project plan is based around existing robotic arms, allowing for quick and low-cost adoption by large and small companies upon completion.
Robot-Assisted Wire Harness Installation
Principal Investigator: QinetiQ North America
Project Description: This project centers on empowering human workers and reducing risk of injury in manufacturing environments by developing a robot-assisted wire harness installation system. This system will augment human cognitive and physical abilities while eliminating errors and focusing on worker assist, rather than worker replacement. The project will engage a twin arm robot to hold, route, and position a wire harness while using eye-safe laser pointers to indicate correct wire to terminal and connector pair locations. In addition to the technology advancement, the project will also feature a workforce development aspect with the creation of a training course for the technology.
Mixed Multi–Angle Robotic IR Camera Control for Thermomechanical Surface Processes
Principal Investigator: Siemens Corporation
Project Description: Current surface processes for aerospace and defense are highly complex, frequently poor yield, and require long and expensive development and qualification processes, all of which contribute to high development, production and deployed system maintenance costs. The project team is working to develop a robotic system for monitoring thermomechanical surface processes of advanced composite materials. By combining in-situ sensing, simulation, and robot motion, the technology can be applied across many applications and industry.
Passive Object Tracking via Multi-Spectra Robotic Sensor Fusion Package and Semantic Segmentation
Principal Investigator: Siemens Corporation
Project Description: This project centers on component tracking and traceability across the entire production line using sensor fusion, computer vision, and robotic and deep learning-based technologies, improving accuracy and lowering cost. The approach is to build a sensor fusion package that will scan factory space and build an accurate image and 3D volumetric map of the space and surfaces in multi-spectra modalities. Additionally, the technology will be designed for flexible integration with existing PLM and factory automation systems, allowing for widespread seamless adoption.
Advanced Robotic Grinding System for Metal Parts
Principal Investigator: Texas A&M University
Project Description: Grinding metal parts remains a repetitive and ergonomically unfriendly task that often generates inconsistent results. This project will empower American workers to transition from manual grinders into safer robot-operated roles through the development of an advanced robotic grinding system. The system will use model-based control and learning algorithms to account for uncertainties in its environment while using automated trajectory generation and 3D part model and identification of surface regions, creating consistent, high-quality results.
Robotic Assistants for Composite Layup
Principal Investigator: University of Southern California
Project Description: Composite layup, despite being labor-intensive, remains a highly manual process and thus limited to low-volume runs or specialty composite item production. The robotic technologies generated through this project aim to make the layup process more collaborative with robot assistants working alongside humans. This will require advancements in perception, planning, and control areas. Additionally, the project will evaluate the quality of robotic layup processes and compare it to the manual process of today.
Automated Wire Harness Assembly
Principal Investigator: Wichita University
Project Description: Today’s wire harness assembly process is a time-intensive, manual process that results in inconsistent product quality and high prices. This project seeks to reduce wire harness assembly costs by using collaborative robots to lay wiring through the development of a unique end effecter. The technology will integrate into existing ECAD, increasing the ease of integration and allowing for adoption by other markets.