The Advanced Robotics for Manufacturing (ARM) Institute announced another eight short-cycle technology projects for funding from its 23-01 Technology Project Call, released earlier this year. Several of the projects are focused on making robotics more scalable in industries where customization and flexibility are key to meeting customer demand with advanced robotics. For robotics to work flexibly on the shop floor, they must interact collaboratively with human operators and machines.
The ARM Institute plans to award nearly $1.56 million in project funding from various sources, for a total contribution of approximately $3.26 million across these eight projects. To date, the ARM Institute has funded and managed more than 150 robotics and artificial-intelligence (AI) technology and workforce development projects.
The 23-01 Project Call asked for proposals to address the following topic areas:
• automated robotic task planning
• multi-robot, multi-human collaboration, task sharing and task allocation
• safe and scalable manufacturing of energetics
• artificial intelligence in robotics for manufacturing
• discovery workshops and market studies.
One-off components: robotics in metal forming
Ohio State University, along with CapSen Robotics, Yaskawa and Robins Air Force Base in Georgia will continue Phase 2 of an autonomous robotics project in metal forming, where there is a growing need for small-volume, high-mix manufacturing. However, the limited supply chain for one-off components and the expensive machining required for complex components works against fast customization. The project seeks to accelerate robotic system productivity to help solve this problem.
Phase 1 of the project was part of an earlier ARM call for projects. In the first project phase, the team designed and deployed an AI robotic system capable of lexibly producing myriad component geometries in a timely and cost-effective manner. Metallic components are commonplace in the commercial automotive sector, high-end auto sports, heavy-duty factory machinery, power plants and in air-, land- and sea-based military equipment.
Phase 2 will accelerate productivity by:
• reducing the time required to position the component on the lower die
• removing the need for pauses during pressing to relieve forces and torques
• allowing for larger amounts of deformation to be taken on each iteration
• reducing the number of images needed for component geometry reconstruction
• reducing the time to heat material to forging temperatures
• reducing the amount of deformation needed to transform the initial geometry to the final geometry
• reducing the frequency of component imaging.
Flexible robotics for precise handling tasks and multiple applications
Siemens and the University of Southern California will undertake a project to replace operators with robots for handling tasks such as scooping and pouring precise amounts without spillage. This is common in the manufacturing of energetic materials and has applications in the pharmaceutical and chemical industries. The team will develop a robotic skill based on AI imitation and reinforcement learning to more safely scoop precise amounts of granular and paste-like materials. The flexibility needed for some operations will lead to reconfigurable robots that can adapt to a different process at lab scale. The two also have a second project in this round to study multi-modal inputs for AI and their potential in manufacturing. This foundation model—like ChatGPT—will dominate the AI landscape. The report will also look at flexible manufacturing and rethinking robotics deployment for one specific purpose.
Garment manufacturing automation: robotic material handling
Apparel Robotics and MassRobotics will do a project aimed at bringing more automation to the garment manufacturing industry. This project will develop new flexible robotic material-handling capabilities, which are required to unload a cutting table or a conveyor that has a number of cut, nested fabric pieces of varying sizes and geometries on it. The project will:
• use a vision system to identify the cut fabric piece based on the shape and other features
• develop an adaptable end-of-arm-tooling (EOAT) to adjust to the geometry of the fabric piece to be picked up
• leverage the fabric gripper technology developed by Apparel Robotics.
The Advanced Robotics for Manufacturing Institute is a Manufacturing Innovation Institute (MII) funded by the Office of the Secretary of Defense under Agreement Number W911NF-17-3-0004 and is part of the Manufacturing USA network. The ARM Institute leverages a diverse ecosystem of nearly 400 consortium members and partners across industry, academia and government to make robotics, autonomy and artificial intelligence more accessible to U.S. manufacturers large and small, train and empower the manufacturing workforce, strengthen the U.S. economy and global competitiveness and elevate national security and resilience.
