MFI is catalyzing the future of manufacturing by combining three key areas: cyberinformation technologies (virtual modeling, big data analytics, augmented reality, Internet of Things, artificial intelligence, cybersecurity, and cloud computing), manufacturing technologies (design, materials, robotics, additive manufacturing, micro/nanofabrication) and the social sciences (learning science, policy, ethics).
1. Current projects in the
Manufacturing Futures Initiative
www.engineering.cmu.edu/mfi/research
2. Distributed learning
for large-scale
multi-robot
path planning
By building a hybrid environment
consisting of numerous simulated and
physical robots, this project will enable
large-scale automation of existing factories
and warehouses.
3. Training in AM using
entertainment and
tutoring technologies
It is impractical for more than three students
and an instructor to use an AM machine at a
given time. Interactive systems that simulate
the operation of AM machines can be used to
enable virtual training of large numbers of
students and users.
4. Robotic
de-powdering
for AM
Conventionally, powders used during AM
processes need to be manually removed
after builds, at risk to human technicians.
These processes can be made safer,
cheaper, and more time-efficient by using
robots equipped with vision sensors and
cleaning actuators.
5. Precise profiling
and protection for
IoT in manufacturing
IoT devices in manufacturing, such as 3D
printers, robotic arms, inspection robots, and
air purifiers, are attractive targets for cyber
attackers. This project aims to provide effective
protection for IoT devices in manufacturing.
6. Transforming AM
of polymers using
FRE-3DP
Only a very limited set of high-performance
polymers can be used in current AM
approaches. By creating a platform
technology based on CMU’s newly developed
FRE-3DP process, this research will enable a
transformative advance in polymers AM.
7. Data-driven fault
detection and
prediction in AM
While automation dramatically boosts
productivity, precision, and efficiency in
manufacturing, a shortcoming is that unforeseen
faults in the system are difficult to diagnose. By
using a reinforcement learning framework,
researchers can take a global view of the
operations of a manufacturing plant to efficiently
diagnose system faults.
8. Profile-3D-printing
for thermally tuned
concrete panels
There is an urgent need to develop new AM
material delivery systems that can combine
multi-material systems, such as concrete with
steel reinforcement, within a single deposition
flow. Through post-processing based on robotic
tooling, researchers can generate topologically
complex shapes optimized for improved
building energy performance.
9. HEALER:
Computationally guided
AM of self-healing
robotic materials
The ability to self-heal is emerging as a highly
desired property in soft robotics, artificial muscles,
and synthetic skins, among other research areas.
AM of elastomeric soft robotic materials with
electrically-actuating compartments can allow
material to be dismembered and reattached,
and the material will self-heal with complete
functional recovery.
10. Learn more about the
Manufacturing Futures Initiative
www.engineering.cmu.edu/mfi/research