Robot chef that can cook any of 2,000 meals at tap of a button to go on sale in 2017

Stirring, adjusting the temperature, pouring and adding ingredients are all basic skills for a chef but they’re slightly harder to achieve for a robot.

However, that’s not the case for this pair of robotic hands, which could be set to revolutionise cooking and kitchen operations.

At present it’s able to knock up a crab bisque, which it creates by replicating the exact movements of a professional chef.

Creator Moley Robotics says that when the commercial version launches in 2017 users will be able to select one of 2,000 dishes from their phone and the robotic hands in the automated kitchen will make it.

If the robot is successful, it could mean we can simply tap a button on our phone to have a meal prepared in time for us coming home from work.

Everything in the automated kitchen has been designed and built from scratch. This allows the hands to be able to pick up and put down utensils, stir food a pan, and then safely turn a hob to the correct temperature.

The robotic hands, which are notoriously difficult to create, use 20 motors, 24 joints and 129 sensors to create the same range of movements that a human hand can make.

The cooking process was recorded in 3D in a special studio where each motion was captured.

The movements of the chef were then transferred into algorithms which could be re-created by the robot.

It isn’t the first robot which has been developed to work in a kitchen but the makers claim it is the first fully automated kitchen set-up to exist.

Other robots that are able to handle the heat of the kitchen include the Cooki, which uses a robotic arm to make meals from pre-portioned ingredients.

Researchers from Cornell University are teaching robots to understand instructions that are given to them by voice rather than a computer programme. So far they have been able to teach one robot how to boil water in a pan.

Meanwhile, in China, Chef Cui has been helping to slice noodles.


Ultimately the creators aim to build an app store for food, which will allow those who purchase the system to download recipes and instructions for the robot.

Mark Oleynik, who founded Moley Robotics, said that it is his aim to use robotics and other technologies that can help to make our lives easier.

“Whether you love food and want to explore different cuisines, or fancy saving a favourite family recipe for everyone to enjoy for years to come, the Automated Kitchen can do this,” said Oleynik.

“It is not just a labour saving device – it is a platform for our creativity. It can even teach us how to become better cooks.”

Images courtesy of Moley Robotics

Global wildlife set to see 67% drop in numbers by 2020: WWF

A landmark report released today by the World Wildlife Fund (WWF) and the Zoological Society of London (ZSL) has found that the world’s vertebrates are set to see an average population drop of 67% from 1970 levels by the end of this decade.

This conclusion was drawn from data collected in The Living Planet Report 2016, the most comprehensive survey of the state of species and ecosystems ever undertaken.

At present, vertebrate populations – that is fish, birds, mammals, amphibians and reptiles – have already seen a 58% population decline since 1970, but the notion of this ramping up to 67% in just four years is deeply concerning.

“For the first time since the demise of the dinosaurs 65 million years ago, we face a global mass extinction of wildlife,” said Mike Barrett, director of science and policy at WWF-UK.

“We ignore the decline of other species at our peril – for they are the barometer that reveals our impact on the world that sustains us. Humanity’s misuse of natural resources is threatening habitats, pushing irreplaceable species to the brink and threatening the stability of our climate.”

The wide-reaching report involved tracking more than 14,000 species of vertebrate from 1970 to 2012; a feat that was achieved using an immense database run by ZSL known as the Living Planet Index.

It draws particular attention to how human activity has impacted on wildlife populations, including through deforestation, overfishing and pollution, as well as the illegal wildlife trade and climate change.

Among the many, many vertebrates affected are Africa’s various species of elephants, which collectively has seen a population drop of 111,000 in the last 10 years, thought to be the result of poaching, leaving only 415,000 left across the entire continent.

Across Asia only 3,900 tigers are now left the wild, as a result of a variety of human activities including habitat destruction and climate change.

And in the waters around Europe, Orcas are seeing a significant decline as a result of the extremely toxic levels of persistent organic pollutants (POPs) found in their blubber.


However despite the concerning news, ZSL and WWF maintain that action can be taken to stop this decimation of the world species, with the wildlife charity calling on members of the public to demonstrate to their governments that they want urgent action to be taken.

“We know how to stop this. It requires governments, businesses and citizens to rethink how we produce, consume, measure success and value the natural environment.,” said Barrett.

“In the UK, this demands a serious plan to strengthen protection for habitats and species and new measures to fast track low-carbon growth. Britain, like all developed nations, must take increasing responsibility for its global footprint.”

“Across land, freshwater and the oceans, human activities are forcing species populations and natural systems to the edge,” added Marco Lambertini, Director General of WWF International. “We have the tools to fix this problem and we need to start using them if we are serious about our own survival and prosperity.”

Near-perfect quantum clones open path to crypto-communication

Physicists have created near-perfect clones of quantum information, their newly developed method allowing them to produce clones that surpass the previous limits of quantum cloning.

A research team from the Australian National University (ANU) and University of Queensland developed the new cloning method using high performance optical amplifiers to clone light encoded with quantum information.

The team’s technique could allow existing fibre optic infrastructure to implement quantum encryption.

“One obstacle to sending quantum information is that the quantum state degrades before reaching its destination. Our cloner has many possible applications, and could help overcome this problem to achieve secure long-distance communication,” said Professor Ping Koy Lam, node director of the ARC Centre of Excellence for Quantum Computation and Communication Technology (CQC2T) at ANU.

Quantum cloning consists of taking an unknown and arbitrary quantum state, which provides the probability distribution for the value of each observable in an isolated quantum system, and making a copy without altering the original state in any way.

The laws of quantum mechanics render it impossible to make a perfect quantum clone, making it necessary for the researchers to use a probabilistic method to demonstrate the possibility of creating clones that exceed the theoretical quality limits. This method builds off work initially proposed by CQC2T researchers, led by Professor Timothy Ralph at the University of Queensland.

“Imagine Olympic archers being able to choose the shots that land closest to the target’s centre to increase their average score,” said Ralph. “By designing our experiment to have probabilistic outputs, we sometimes ‘get lucky’ and recover more information than is possible using existing deterministic cloning methods. We use the results closest to a ‘bullseye’ and discard the rest.”

Images courtesy of Lee Henderson/UNSW

A visual representation of the cloning process. Images courtesy of Lee Henderson/UNSW

The new method is capable of generating quantum clones with a success rate of roughly 5%, a higher quality than has ever been created before. The probabilistic technique allows for the creation of up to five clones of a single quantum state and works by first encoding information onto a light beam in a fragile quantum state.

“At the heart of the demonstration is a ‘noiseless optical amplifier’. When the amplification is good enough, we can then split a light beam into clones,” said Ralph. “’Amplify-then-split’ allows us to clone the light beam with minimal distortion, so that it can still be read with exquisite precision.”

The importance of quantum cloning lies in opening up valuable experimental possibilities, as well as crypto-communication. There is a technological race to make use of quantum information for ultra-secure encryption, currently restricted by its limited communication range.

The Australian research team hopes, however, that their new method may open up the path to impenetrable encryption on communications between two parties.

The research is detailed today in the journal Nature Communications.

AI judge comes to the same conclusions as humans in human rights trials

Artificial intelligence has been used to predict the outcome of judicial decisions made by the European Court of Human Rights (ECtHR) with 79% accuracy.

Researchers at University College London and the universities of Sheffield and Pennsylvania fed the AI English language data sets pertaining to 584 cases and let the algorithm find patterns in the text.

All of the cases related to specific articles of the Convention on Human Rights (Articles 3, 6 and 8). To prevent bias and mislearning, they selected an equal number of violation and non-violation cases.

“We don’t see AI replacing judges or lawyers, but we think they’d find it useful for rapidly identifying patterns in cases that lead to certain outcomes. It could also be a valuable tool for highlighting which cases are most likely to be violations of the European Convention on Human Rights,” explained Dr Nikolaos Aletras, who led the study at UCL Computer Science.

Image courtesy of Oleg Mikhaylov /

Image courtesy of Oleg Mikhaylov /

In developing the AI, the researchers found that judgements by the ECtHR are often based on non-legal facts rather than directly legal arguments, which makes the job of an AI even harder.

To get to a judgement, the AI deliberated over language used as well as the topics and circumstances mentioned in the case text. The AI also looked at the factual background to the case.

By combining the information extracted from the case files, the AI achieved an accuracy of 79%.

“Previous studies have predicted outcomes based on the nature of the crime, or the policy position of each judge, so this is the first time judgements have been predicted using analysis of text prepared by the court. We expect this sort of tool would improve efficiencies of high level, in demand courts, but to become a reality, we need to test it against more articles and the case data submitted to the court,” added Dr Lampos.


The articles chosen by the researchers were used because they represented cases about fundamental rights and because there was a large amount of published data on them.

However, the researchers would have like to have been given access to application made directly to court, rather than having to make do with the outcome of those applications.

“Ideally, we’d test and refine our algorithm using the applications made to the court rather than the published judgements, but without access to that data we rely on the court-published summaries of these submissions,” explained co-author, Dr Vasileios Lampos, UCL Computer Science.