[ad_1]
Atlantic University of FloridaThe College of Engineering and Computer Science and Institute for Sensing and Embedded Network Systems Engineering (I-SENSE) is expected to receive a $ 1 million grant from the National Science Foundation (NSF) to develop the country’s first test bed platform that connects robots using extremely fast millimeter wave links.
The new generation of networked cooperating robots will be developed by researchers at the university’s Center for Connected Autonomy and Artificial Intelligence (ca-ai.fau.edu). Rather than using traditional methods such as Wi-FI and 5G, the FAU research team will equip a team of robots with individual programmable millimeter wave modems to create their own very high-speed autonomous network, then train them to perform tasks as a team. .
In addition, the autonomous mm-wave network will have dual use in robot-to-robot communication and detection. Researchers at the Center will develop new multi-agent learning algorithms executed on networked robots, as well as protocols for the operation of networked robotic teams.
“Just like humans, autonomous robots must communicate with each other to learn together and accomplish a team task or mission such as search and rescue,†said Dimitris Pados, Ph.D., Principal Investigator, Schmidt Eminent Scholar Professor, Department of Electrical Engineering and Computer Science, FAU I-SENSE Fellow and Director of the Center for Connected Autonomy and Artificial Intelligence, which has already collected more than $ 9 million federal funding since its official opening in March. “Our five test bench robots will be able to communicate at ultra-high speeds of gigabits per second by forming and directing ‘beams’ towards each other, which will also allow them to see through objects according to their needs. They’ll see, so to speak, what other bots are picking up in real time, resulting in five times as many eyes thanks to the near instantaneous exchange of large volumes of data.
The FAU noted that the connected millimeter wave robotic platform for learning and team AI operations consists of five mobile robotic modules on the ground, all equipped with on-board mmWave programmable transceivers, LiDAR and GPU (graphics processing unit). Two of the mobile robotic modules are equipped with a robotic arm. Additional major platform components include two programmable fixed point mmWave transceivers and a base station GPU tower.
“By recreating the mobility and dynamism seen in real-world scenarios, our project will be the first to enable an in-depth experimental evaluation of multi-function mmWave radios and connected AI systems, and creates an opportunity to strengthen the ongoing partnerships between the mmWave network and the academic robotics and industry communities that have led the development of mmWave and subTHz software-defined radios and robotic platforms, â€said George sklivanite, Ph.D., co-PI, Research Assistant Professor, Department of Electrical Engineering and Computer Science, and member of FAU I-SENSE, which works with Pados and co-PI Xiangnan Zhong, Ph.D., assistant professor in the Department of Electrical and Computer Engineering; and Jason hallstrom, Ph.D., director of FAU I-SENSE and professor in the Department of Electrical and Computer Engineering.
To train the robots, FAU will use organized data sets developed as part of their ACE (Autonomous Conformity Evaluation of Tensor Data by Means of Novel L1-norm Principal-Component Analysis) project funded by the United States Air Force Office of Scientific Research. , which aims to assess data quality and remove erroneous entries. They are also working to stop attempts by saboteurs to inject faulty data to trick autonomous systems by performing real-time operational data monitoring.
“Anytime you have bad robots or a bad autonomous system, it’s usually the data that’s causing the problem,†Pados continued. “For example, a somewhat defective sensor that produces the wrong measurement at the wrong time may be enough to cause a catastrophic event. We are developing new theories and algorithms to identify data values ​​that do not conform or make sense when correlated with each other, to serve as a warning system to notify humans that some thing is wrong. We do this both for the training phase of AI systems as well as for their operational phase afterwards. “
The new connected robotics platform will also provide a unique research and training opportunity at the intersection of mmWave wireless networks, robotics and multi-agent AI. Researchers from all US universities will be offered remote access to the FAU platform for experimentation.
“This new platform developed by our team at the Center for Connected Autonomy and Artificial Intelligence will advance research activities in the area of ​​multi-agent AI, as well as mmWave networks and communications,” added Stella Batalama, Ph.D., Dean, College of Engineering and Computer Science. “This will enable rapid testing and reproducible comparable assessment of collaborative AI operations and distributed detection, positioning, synchronization, navigation and communication developed by researchers from different institutions.”
Related
[ad_2]
