Thursday, November 16, 2017

CMU to participate in LoRaWAN Academy curriculum

The LoRaWAN Academy announced the participation of the College of Engineering at Carnegie Mellon University in the LoRaWAN Academy curriculum yesterday, where CMU had developed its own network, OpenChirp, an LPWAN network.
“Connecting sensors is often the most expensive and challenging part of a deployment especially when they are located in remote areas where data needs to travel long distances. By implementing battery-operated, low-powered LoRaWAN-based devices and the LoRaWAN protocol, OpenChirp demonstrates that it is feasible to scale low-powered sensing devices for use across large areas, like campuses, manufacturing plants or even cities,” said Anthony Rowe, an associate professor of electrical and computer engineering, who leads the OpenChirp project at CMU. “At Carnegie Mellon, students are using OpenChirp to develop IoT applications including smart grid demand / response, air quality sensing, and a campus asset-tracking system.”
Carnegie Mellon’s work is a prime example of the LoRaWAN-based projects that students can develop with LoRa Technology and the LoRaWAN open protocol. See Market Insider.

Monday, November 13, 2017

UC Berkeley Professor and TerraSwarm PI Pieter Abbeel heads new startup of AI hotshots

Professor Peter Abbeel's research at UC Berkeley's Robot Lab had been working on adaptive learning through demonstration, initially with the famous towel-folding performed by a robot and the robotic knot-tying from 2013. This week, Abbeel and several colleagues announced a new startup (with US $7 million in seed funding) called Embodied Intelligence which will “enable industrial robot arms to perceive and act like humans instead of just strictly following pre-programmed trajectories.”

The new project draws on the groundwork laid by BRETT, a robot that represented a deep reinforcement learning breakthrough for UC Berkeley in 2015. The acronym, believe it or not, stands for Berkeley Robot for the Elimination of Tedious Tasks.

“Right now, if you want to set up a robot, you program that robot to do what you want it to do, which takes a lot of time and a lot of expertise,” said Abeel. “With our advances in machine learning, we can write a piece of software once — machine learning code that enables the robot to learn — and then when the robot needs to be equipped with a new skill, we simply provide new data.”

The new start-up plans to make robots more skillful and quicker to learn, which could have an enormous impact on manufacturing.

“The goal is to bring the cutting-edge research to robotics and manufacturing,” Abbeel says. He says the techniques his company is developing will enable robots to do a range of things that are currently too time-consuming to be programmed in.
See articles in IEEE Spectrum, Electronics 360 and MIT Technology Review for more information.