PRESS-NEWS.org - Press Release Distribution
PRESS RELEASES DISTRIBUTION

Animal brain inspired AI game changer for autonomous robots

First neuromorphic vision and control of a flying drone

Animal brain inspired AI game changer for autonomous robots
2024-05-15
(Press-News.org)

A team of researchers at Delft University of Technology has developed a drone that flies autonomously using neuromorphic image processing and control based on the workings of animal brains.  Animal brains use less data and energy compared to current deep neural networks running on GPUs (graphic chips). Neuromorphic processors are therefore very suitable for small drones because they don’t need heavy and large hardware and batteries. The results are extraordinary: during flight the drone’s deep neural network  processes data up to 64 times faster and consumes three times less energy than when running on a GPU. Further developments of this technology may enable the leap for drones to become as small, agile, and smart as flying insects or birds. The findings were recently published in Science Robotics.

Learning from animal brains: spiking neural networks

Artificial intelligence holds great potential to provide autonomous robots with the intelligence needed for real-world applications. However, current AI relies on deep neural networks that require substantial computing power. The processors made for running deep neural networks (Graphics Processing Units, GPUs) consume a substantial amount of energy. Especially for small robots like flying drones this is a problem, since they can only carry very limited resources in terms of sensing and computing.

Animal brains process information in a way that is very different from the neural networks running on GPUs. Biological neurons process information asynchronously, and mostly communicate via electrical pulses called spikes. Since sending such spikes costs energy, the brain minimizes spiking, leading to sparse processing.

Inspired by these properties of animal brains, scientists and tech companies are developing new, neuromorphic processors. These new processors allow to run spiking neural networks and promise to be much faster and more energy efficient.

“The calculations performed by spiking neural networks are much simpler than those in standard deep neural networks.”, says Jesse Hagenaars, PhD candidate and one of the authors of the article, “Whereas digital spiking neurons only need to add integers, standard neurons have to multiply and add floating point numbers. This makes spiking neural networks quicker and more energy efficient. To understand why, think of how humans also find it much easier to calculate 5 + 8 than to calculate 6.25 x 3.45 + 4.05 x 3.45.”

This energy efficiency is further boosted if neuromorphic processors are used in combination with neuromorphic sensors, like neuromorphic cameras. Such cameras do not make images at a fixed time interval. Instead, each pixel only sends a signal when it becomes brighter or darker. The advantages of such cameras are that they can perceive motion much more quickly, are more energy efficient, and function well both in dark and bright environments. Moreover, the signals from neuromorphic cameras can feed directly into spiking neural networks running on neuromorphic processors. Together, they can form a huge enabler for autonomous robots, especially small, agile robots like flying drones.

First neuromorphic vision and control of a flying drone

In an article published in Science Robotics on May 15, 2024, researchers from Delft University of Technology, the Netherlands, demonstrate for the first time a drone that uses neuromorphic vision and control for autonomous flight. Specifically, they developed a spiking neural network that processes the signals from a neuromorphic camera and outputs control commands that determine the drone’s pose and thrust. They deployed this network on a neuromorphic processor, Intel’s Loihi neuromorphic research chip, on board of a drone. Thanks to the network, the drone can perceive and control its own motion in all directions.

“We faced many challenges,” says Federico Paredes-Vallés, one of the researchers that worked on the study, “but the hardest one was to imagine how we could train a spiking neural network so that training would be both sufficiently fast and the trained network would function well on the real robot. In the end, we designed a network consisting of two modules. The first module learns to visually perceive motion from the signals of a moving neuromorphic camera. It does so completely by itself, in a self-supervised way, based only on the data from the camera. This is similar to how also animals learn to perceive the world by themselves. The second module learns to map the estimated motion to control commands, in a simulator. This learning relied on an artificial evolution in simulation, in which networks that were better in controlling the drone had a higher chance of producing offspring. Over the generations of the artificial evolution, the spiking neural networks got increasingly good at control, and were finally able to fly in any direction at different speeds. We trained both modules and developed a way with which we could merge them together. We were happy to see that the merged network immediately worked well on the real robot.”

With its neuromorphic vision and control, the drone is able to fly at different speeds under varying light conditions, from dark to bright. It can even fly with flickering lights, which make the pixels in the neuromorphic camera send great numbers of signals to the network that are unrelated to motion.

Improved energy efficiency and speed by neuromorphic AI

“Importantly, our measurements confirm the potential of neuromorphic AI. The network runs on average between 274 and 1600 times per second. If we run the same network on a small, embedded GPU, it runs on average only 25 times per second, a difference of a factor ~10-64! Moreover, when running the network, , Intel’s Loihi neuromorphic research chip consumes 1.007 watts, of which 1 watt is the idle power that the processor spends just when turning on the chip. Running the network itself only costs 7 milliwatts. In comparison, when running the same network, the embedded GPU consumes 3 watts, of which 1 watt is idle power and 2 watts are spent for running the network. The neuromorphic approach results in AI that runs faster and more efficiently, allowing deployment on much smaller autonomous robots.”, says Stein Stroobants, PhD candidate in the field of neuromorphic drones.

Future applications of neuromorphic AI for tiny robots

“Neuromorphic AI will enable all autonomous robots to be more intelligent,” says Guido de Croon, Professor in bio-inspired drones, “but it is an absolute enabler for tiny autonomous robots. At Delft University of Technology’s Faculty of Aerospace Engineering, we work on tiny autonomous drones which can be used for applications ranging from monitoring crop in greenhouses to keeping track of stock in warehouses. The advantages of tiny drones are that they are very safe and can navigate in narrow environments like in between ranges of tomato plants. Moreover, they can be very cheap, so that they can be deployed in swarms. This is useful for more quickly covering an area, as we have shown in exploration and gas source localization settings.”

“The current work is a great step in this direction. However, the realization of these applications will depend on further scaling down the neuromorphic hardware and expanding the capabilities towards more complex tasks such as navigation.”

END


[Attachments] See images for this press release:
Animal brain inspired AI game changer for autonomous robots Animal brain inspired AI game changer for autonomous robots 2 Animal brain inspired AI game changer for autonomous robots 3

ELSE PRESS RELEASES FROM THIS DATE:

Summers warm up faster than winters, fossil shells from Antwerp show

Summers warm up faster than winters, fossil shells from Antwerp show
2024-05-15
Niels de Winter, affiliated with the Department of Earth Sciences at Vrije Universiteit Amsterdam and the AMGC research group at Vrije Universiteit Brussel, measured alongside colleagues from institutions such as the Institute for Natural Sciences in Brussels the chemical composition of fossil shells from Antwerp, Belgium. Those shells originate from molluscs such as oysters, cockles, and scallops found during the construction works of the Kieldrecht Lock. The molluscs lived lived during the Pliocene, approximately ...

Wearing face masks did not reduce risk of COVID infection after first Omicron wave, research shows 

2024-05-15
  After the first Omicron wave, research shows that many of the risks of Covid infection changed  Before February 2022, always wearing face masks and being retired were associated with reduced risk, but not after  Overseas travel was not associated with increased risk prior to February 2022, but then became a significant risk  Peer reviewed – meta-regression- humans  New research from the University of East Anglia has found that wearing face masks did not lower the risk of Covid infection following the initial surge of the Omicron variant.  The analysis of official data found that several risk ...

SF State receives $14M from the Genentech Foundation to support underrepresented students in STEM

2024-05-15
SAN FRANCISCO – May 15, 2024 – San Francisco State University announced today that it received $14 million from the Genentech Foundation to support two University programs that are training the next generation of life sciences leaders. The new five-year grant is the latest in the Genentech Foundation’s transformational support for University programs, which has totaled more than $33 million during their long-lasting partnership. This partnership has impacted more than 700 students since 2008, and an additional 350 students are projected to be supported by the new funding. The new funds will continue sponsoring San Francisco State’s Genentech ...

Penalties for dropping out of ecosystem services incentive programs should equal lost environmental benefits

2024-05-15
Payment for Ecosystem Services programs (PES) are important tools that governments around the world use to improve water quality, protect forests and wildlife habitat, and sequester carbon. Under these programs, landowners - usually farmers - are paid to use their land in ways that protect or restore the environment, such as replacing row crops with trees or grassy zones adjacent to waterways. Many PES program contracts last 5 to 20 years, but participant drop out rates have consistently risen over the years. A recent study by University of Maryland economists showed that PES programs ...

Lithuanian researchers’ new development in solar cell technology – a promise of a significant advancement in the field

Lithuanian researchers’ new development in solar cell technology – a promise of a significant advancement in the field
2024-05-15
Researchers from Kaunas University of Technology (KTU), Lithuania, who contributed to the development of record-breaking solar cells a few years ago, expanded their invention. The self-assembled monolayers can now be applied not only in inverted but also in regular structure perovskite solar cells. Self-assembling molecules arrange themselves into a single-molecule-thick layer and in this case, they act as an electron-transporting layer in solar cells.  “The molecules that make up these monolayers, like a clever glue, ...

A simple internet with significant possibilities

A simple internet with significant possibilities
2024-05-15
It’s one thing to dream up a quantum internet that could send hacker-proof information around the world via photons superimposed in different quantum states. It’s quite another to physically show it’s possible. That’s exactly what Harvard physicists have done, using existing Boston-area telecommunication fiber, in a demonstration of the world’s longest fiber distance between two quantum memory nodes to date. Think of it as a simple, closed internet between point A and B, carrying a signal encoded not by classical ...

Unwrapping the origin story of the baobab

Unwrapping the origin story of the baobab
2024-05-15
The baobab (Adansonia) is a genus of trees with eight extant (in existence currently) species and a long history of humans marveling at them. For as much admiration the baobabs get, there is an equal amount of mystery surrounding their origin. Genomic and ecological analyses recently done by a global research team led by Sino-Africa Joint Research Center, CAS (hosted by Wuhan Botanical Garden of the Chinese Academy of Sciences), suggest that Madagascar, is the origin from where all other baobab species hail. With a deeper understanding of the baobabs' genetics, researchers are hoping to uncover some clues on what ...

The origin and long-distance travels of upside down trees

The origin and long-distance travels of upside down trees
2024-05-15
The iconic baobabs, also known as upside-down trees, or the tree of life, have much cultural significance, inspiring innumerable arts, folklore, and traditions. A research published in Nature, involving international collaboration between Wuhan Botanical Garden (China), Royal Botanic Gardens (Kew, UK), University of Antananarivo (Madagascar) and Queen Mary University of London (UK) reveal a remarkable example of species radiation in Madagascar followed by long distance dispersal to Africa and Australia. With speciation, an astonishing divergence of pollination mechanisms evolved, that exploit hawkmoths, bats and lemurs for ...

Some mice may owe their monogamy to a newly evolved type of cell

Some mice may owe their monogamy to a newly evolved type of cell
2024-05-15
NEW YORK, NY — What makes the oldfield mouse steadfastly monogamous throughout its life while its closest rodent relatives are promiscuous? The answer may be a previously unknown hormone-generating cell, according to a new study published online today in Nature from scientists at Columbia's Zuckerman Institute. "The hormone from these cells was actually first discovered in humans many decades ago, but nobody really knew what it did," said Andrés Bendesky, MD, PhD, a principal investigator at Columbia's Zuckerman Institute. "We’ve discovered ...

Mortality in patients hospitalized for COVID-19 vs influenza in fall-winter 2023-2024

2024-05-15
About The Study: This study found that in fall-winter 2023-2024, the risk of death in patients hospitalized for COVID-19 was greater than the risk of death in patients hospitalized for seasonal influenza.  Corresponding Author: To contact the corresponding author, Ziyad Al-Aly, M.D., email ziyad.alaly@va.gov. To access the embargoed study: Visit our For The Media website at this link https://media.jamanetwork.com/ (doi:10.1001/jama.2024.7395) Editor’s Note: Please see the article for additional information, including other authors, author contributions and affiliations, conflict of interest and financial disclosures, and funding and support. #  #  ...

LAST 30 PRESS RELEASES:

Nucleoporin93: A silent protector in vascular health

Can we avert the looming food crisis of climate change?

Alcohol use and antiobesity medication treatment

Study reveals cause of common cancer immunotherapy side effect

New era in amphibian biology

Harbor service, VAST Data provide boost for NCSA systems

New prognostic model enhances survival prediction in liver failure

China focuses on improving air quality via the coordinated control of fine particles and ozone

Machine learning reveals behaviors linked with early Alzheimer’s, points to new treatments

Novel gene therapy trial for sickle cell disease launches

Engineering hypoallergenic cats

Microwave-induced pyrolysis: A promising solution for recycling electric cables

Cooling with light: Exploring optical cooling in semiconductor quantum dots

Breakthrough in clean energy: Scientists pioneer novel heat-to-electricity conversion

Study finds opposing effects of short-term and continuous noise on western bluebird parental care

Quantifying disease impact and overcoming practical treatment barriers for primary progressive aphasia

Sports betting and financial market data show how people misinterpret new information in predictable ways

Long COVID brain fog linked to lung function

Concussions slow brain activity of high school football players

Study details how cancer cells fend off starvation and death from chemotherapy

Transformation of UN SDGs only way forward for sustainable development 

New study reveals genetic drivers of early onset type 2 diabetes in South Asians 

Delay and pay: Tipping point costs quadruple after waiting

Magnetic tornado is stirring up the haze at Jupiter's poles

Cancers grow uniformly throughout their mass

Researchers show complex relationship between Arctic warming and Arctic dust

Brain test shows that crabs process pain

Social fish with low status are so stressed out it impacts their brains

Predicting the weather: New meteorology estimation method aids building efficiency

Inside the ‘swat team’ – how insects react to virtual reality gaming 

[Press-News.org] Animal brain inspired AI game changer for autonomous robots
First neuromorphic vision and control of a flying drone