Category: Allgemein

Der GPS-Tracker Protzhelis Meintal meldet Pferdeahänger-Diebstahl

www.auto-bussgeld-opfer.de

MIT researchers have developed an automated way to design
customized hardware, or “brains,” that speeds up a robot’s
operation.  Image: Jose-Luis Olivares, MIT

Contemporary robots can move quickly. “The motors are fast,
and they’re powerful,” says Sabrina Neuman.

Yet in complex situations, like interactions with people, robots
often don’t move quickly. “The hang up is what’s going on in
the robot’s head,” she adds.

Perceiving stimuli and calculating a response takes a
“boatload of computation,” which limits reaction time, says
Neuman, who recently graduated with a PhD from the MIT Computer
Science and Artificial Intelligence Laboratory (CSAIL). Neuman has
found a way to fight this mismatch between a robot’s “mind”
and body. The method, called robomorphic computing, uses a
robot’s physical layout and intended applications to generate a
customized computer chip that minimizes the robot’s response
time.

The advance could fuel a variety of robotics applications,
including, potentially, frontline medical care of contagious
patients. “It would be fantastic if we could have robots that
could help reduce risk for patients and hospital workers,” says
Neuman.

Neuman will present the research at this April’s International
Conference on Architectural Support for Programming Languages and
Operating Systems. MIT co-authors include graduate student Thomas
Bourgeat and Srini Devadas, the Edwin Sibley Webster Professor of
Electrical Engineering and Neuman’s PhD advisor. Other co-authors
include Brian Plancher, Thierry Tambe, and Vijay Janapa Reddi, all
of Harvard University. Neuman is now a postdoctoral NSF Computing
Innovation Fellow at Harvard’s School of Engineering and Applied
Sciences.

There are three main steps in a robot’s operation, according
to Neuman. The first is perception, which includes gathering data
using sensors or cameras. The second is mapping and localization:
“Based on what they’ve seen, they have to construct a map of
the world around them and then localize themselves within that
map,” says Neuman. The third step is motion planning and control
— in other words, plotting a course of action.

These steps can take time and an awful lot of computing power.
“For robots to be deployed into the field and safely operate in
dynamic environments around humans, they need to be able to think
and react very quickly,” says Plancher. “Current algorithms
cannot be run on current CPU hardware fast enough.”

Neuman adds that researchers have been investigating better
algorithms, but she thinks software improvements alone aren’t the
answer. “What’s relatively new is the idea that you might also
explore better hardware.” That means moving beyond a
standard-issue CPU processing chip that comprises a robot’s brain
— with the help of hardware acceleration.

Hardware acceleration refers to the use of a specialized
hardware unit to perform certain computing tasks more efficiently.
A commonly used hardware accelerator is the graphics processing
unit (GPU), a chip specialized for parallel processing. These
devices are handy for graphics because their parallel structure
allows them to simultaneously process thousands of pixels. “A GPU
is not the best at everything, but it’s the best at what it’s
built for,” says Neuman. “You get higher performance for a
particular application.” Most robots are designed with an
intended set of applications and could therefore benefit from
hardware acceleration. That’s why Neuman’s team developed
robomorphic computing.

The system creates a customized hardware design to best serve a
particular robot’s computing needs. The user inputs the
parameters of a robot, like its limb layout and how its various
joints can move. Neuman’s system translates these physical
properties into mathematical matrices. These matrices are
“sparse,” meaning they contain many zero values that roughly
correspond to movements that are impossible given a robot’s
particular anatomy. (Similarly, your arm’s movements are limited
because it can only bend at certain joints — it’s not an
infinitely pliable spaghetti noodle.)

The system then designs a hardware architecture specialized to
run calculations only on the non-zero values in the matrices. The
resulting chip design is therefore tailored to maximize efficiency
for the robot’s computing needs. And that customization paid off
in testing.

Hardware architecture designed using this method for a
particular application outperformed off-the-shelf CPU and GPU
units. While Neuman’s team didn’t fabricate a specialized chip
from scratch, they programmed a customizable field-programmable
gate array (FPGA) chip according to their system’s suggestions.
Despite operating at a slower clock rate, that chip performed eight
times faster than the CPU and 86 times faster than the GPU.

“I was thrilled with those results,” says Neuman. “Even
though we were hamstrung by the lower clock speed, we made up for
it by just being more efficient.”

Plancher sees widespread potential for robomorphic computing.
“Ideally we can eventually fabricate a custom motion-planning
chip for every robot, allowing them to quickly compute safe and
efficient motions,” he says. “I wouldn’t be surprised if 20
years from now every robot had a handful of custom computer chips
powering it, and this could be one of them.” Neuman adds that
robomorphic computing might allow robots to relieve humans of risk
in a range of settings, such as caring for covid-19 patients or
manipulating heavy objects.

“This work is exciting because it shows how specialized
circuit designs can be used to accelerate a core component of robot
control,” says Robin Deits, a robotics engineer at Boston
Dynamics who was not involved in the research. “Software
performance is crucial for robotics because the real world never
waits around for the robot to finish thinking.” He adds that
Neuman’s advance could enable robots to think faster,
“unlocking exciting behaviors that previously would be too
computationally difficult.”

Neuman next plans to automate the entire system of robomorphic
computing. Users will simply drag and drop their robot’s
parameters, and “out the other end comes the hardware
description. I think that’s the thing that’ll push it over the
edge and make it really useful.”

This research was funded by the National Science Foundation, the
Computing Research Agency, the CIFellows Project, and the Defense
Advanced Research Projects Agency.

Originally published by
Daniel Ackerman | MIT News
Office | January 21, 2021
MIT

Wer beim Treibstoffkauf nicht zu viel zahlen möchte, sollte sich vorab informieren. (Bildquelle: bunyarit/iStock)

A sensor network powered by an artificial intelligence (AI)
algorithm developed by scientists fromNanyang
Technological University, Singapore (NTU Singapore) can
accurately detect, in real-time, gas leaks and unwanted water
seepage into gas pipeline networks. 

 

Successful in field trials conducted on Singapore’s gas
pipeline networks, the algorithm has been patented and spun off
into a start-up named Vigti, which is now
commercialising the technology. It has recently raised early
start-up funding from Artesian Capital and Brinc, Hong Kong.

 

The NTU start-up is incubated by the
University’s EcoLabs Centre of Innovation for
Energy, a national centre launched in April 2019 to help
small and medium-sized enterprises (SMEs) and start-ups innovate,
grow and thrive in the energy sector.

 

A smart warning system that can detect gas leaks and broken
gas pipes in real-time has been a long-term goal for the public
utility industry, as the current industry best practice for
inspecting pipes is for workers to undertake manual surveillance at
regular intervals. 

 

While big leaks can be easily detected via conventional
sensors as the gas volume and pressure differences will fluctuate
sharply in the pipe networks, small leaks are much harder to
detect.

 

In 2014, the Energy Market Authority of Singapore (EMA)
awarded a grant to NTU researchers led by Dr Justin
Dauwels, then an associate professor at the School of
Electrical & Electronic Engineering, to develop an anomaly
identification software for low-pressure pipeline networks.

 

Over a four-year period starting from 2015, the NTU
researchers developed, deployed and tested their AI solution on
certain segments of the local city gas network in Singapore over
six months, which was shown to be successful in detecting all
tested types of anomalies.

 

“We have designed novel AI algorithms, trained on a massive
amount of field data, to identify anomalies such as leaks, bursts
and water ingress, which can aid energy companies to better manage
their pipe networks,” added Dr Dauwels, who is now the AI Advisor
of Vigti.

 

The EMA funded project concluded in 2019 after the successful
field trials and Vigti was then formed to continue developing the
innovation and bring it to the global market.

 

Chief Executive Officer of Vigti, Mr Ishaan
Gupta, said: “We aim to reduce the methane emissions in
the global gas supply chain to a minimum, with our early detection
system, helping companies to save costs while protecting lives. Our
mission is to create a safe, smart and a sustainable world, one
pipeline at a time.”

 

Professor Subodh Mhaisalkar, Executive Director of the
Energy Research Institute @ NTU (ERIAN) and
a Governing Board Member of EcoLabs, said
Vigti’s technology is a prime example of an NTU innovation going
from lab to market.

 

“With ageing infrastructure and rising gas leaks around the
world, Vigti’s solution is well-positioned to solve a global
problem, mitigating gas emissions and leaks that impact climate
change and pose a potential threat to the well-being of
communities. At NTU EcoLabs, we have pooled together expertise and
the funding for Vigti, which enabled the pilot-scale testing of the
technology, paving the way for actual market adoption.”

 

 

While within a typical gas network there are sensors installed
at regulator points that can detect major fluctuation in the
network and calculate the Unaccounted-for-Gas (UFG) loss, small
leaks and cracks can escape notice and thus must be manually
detected.

 

With the conventional threshold-based approach, leaks can only
be detected if the pressure drop due to the leak is higher than the
pressure variation of the network during normal operation. If it is
lower than the pressure variation, the leaks will be very hard to
detect unless the pipes are inspected manually. 

 

The cumulative loss of all the small leaks for major companies
across the world is estimated between 1.5 to 3 per cent of total
gas consumption. 

 

Total natural gas consumption worldwide is estimated to be 3.9
trillion cubic metres as of 2019, thus even a 1 per cent loss would
mean some 39 billion cubic metres globally (10 times the total
consumption of natural gas of Singapore in 2017).

 

 

To tackle these issues, the NTU team performed various
computational simulations to understand the leak and water ingress
phenomena in the city’s natural gas distribution networks.

 

A variety of sensors that can measure pressure, flow,
temperature and vibration, were deployed and the resulting signals
associated with the anomalies in the network’s pipes were
analysed. This process established unique ‘signatures’ within
the sensor data for each anomaly.

 

Using machine learning and AI, the team then developed a
software algorithm that is extremely sensitive in detecting
anomalies by matching these unique signatures within the sensor
data that is routinely monitored.  

 

During the field trial, a total of 16 pressure sensors and 4
flow sensors of various types were deployed at the riser, service
line and main line, across three different locations. Data was then
analysed at each location and leak and water ingress tests were
also performed at these sites. 

 

At the end of the project, a test was done to establish the
effectiveness of NTU’s AI comprising 13 different anomaly tests.
All 13 were successfully identified by the algorithm as leaks,
along with the nearest sensor location and the time duration of
these leaks. 

 

Originally published by

XZENT X-427: 2-DIN car radio and navigation

https://www.diesel-auto-opfer.de/