Share |

Getting on the right SIDE

The Ambient Kitchen is situated in the Culture Lab, Newcaslte University. It's designed to allow researchers to analyse three areas: Activity monitoring: tracking manipulation of objects; Nutrition advisor: based on the food items taken out of the fridge and cupboards and Sequence prompting: voice prompts and small lights on objects are used to help people with dementia. In the video you will also see a researcher holding specially designed kitchen utensils with accelerometers inside. Image courtesy SIDE project.

Most people have heard of smartphones, but what about the ‘smart kitchen’? Patrick Olivier, a human-computer interaction researcher from Newcastle University, in the UK, is merging the kitchen with the laboratory in the Ambient Kitchen, which he will be presenting at the Digital Engagement 2011 conference in Newcastle University in mid-November.

It uses the latest digital technology to help people in their own kitchens - for example, those living with dementia.

“When you ask older people what they want, they’re adamant to live in their own homes for as long as possible. But, neurodegenerative conditions such as dementia affect many people as they age,” said Paul Watson, a computer scientist from Newcastle University and director of the Social Inclusion through the Digital Economy (SIDE) initiative.

“These conditions severely impact their ability to do relatively complex activities like cooking or even making a cup of tea. Even basic kitchen activities require humans to manage quite long sequences of actions,” said Watson.

Accelerometers in kitchen objects

The Ambient Kitchen augments a kitchen with state-of-art sensors and communication equipment, using a combination of accelerometers, RFID (Radio-frequency identification technology to track objects with radio waves) sensors embedded in worktops and utensils.

“A key aspect of our work is not to create a new environment, but to invisibly embed the technology, so for example, we used a 3D printer to create a new handle for the kitchen utensils so we could insert an accelerometer inside it to track the object,” Watson said.  

“You can’t put an RFID in a carrot, and people will not accept a camera in the kitchen, so we created a chopping board with fiber optic sensors that act like a low-resolution camera and can recognize most foods that are placed on it: fresh or frozen to meat or vegetable. The Ambient Kitchen will eventually provide a ‘context aware’ computing environment that will be able to help people in many ways,” said Olivier.    

“Why are you still holding that spoon in mid-air?”

These devices and technologies are integrated into an automatic care system called SNAP (SyNdetic Assistance Processes). It is a software application that can keep track of up to 25 utensils simultaneously and data about the actions and intention of a person in the kitchen.

SNAP makes a kitchen smarter by understanding what a person is doing with each object. When a knife is being used, the recognition software understands if it is being held, or used to dice or slice. The system will prompt someone if they stop moving the object for a prolonged period of time.

The researchers plan to install an Ambient Kitchen in a dementia care home in Gateshead, UK, next year. After that, “our project is a research platform to enable people to connect more with their homes. It will eventually provide on-demand access to resources for users in their own homes,” said Olivier.

Learn French and improve your cooking

This video shows you how the Ambient Kitchen provides audio and visual cues to help a person cook a delicious meal. Image courtesy SIDE project

The kitchen is being developed to serve other purposes as well, such as teaching catering students French by associating duties in the kitchen with words in the French language. The kitchen tracks whether students have completed a step or have made a mistake in a recipe. Learning a language while engaged in a real world task is known to be effective according to the researchers, but this is the first time it has been supported using digital technology.

The researchers are also working with technology company Philips on the Balance@Home project to improve a person’s cooking competence.

While data from the Ambient Kitchen is currently being stored on a local machine with data output from all utensils at a total of 15 megabytes-per-second, a prototype allows data to be sent and stored in a new cloud platform called e-Science Central.

The e-Science Central cloud

“It’s a web-based system, written in Java, that provides a number of data analysis and storage applications. You can upload sensor data or code directly for analysis and processing. Our platform is a middleware for the clouds, which runs on top of Microsoft Azure, Amazon EC2 and private clouds. When supporting our projects, the e-Science Central system uses up to 200 computing cores simultaneously. We have a commercially available version of the e-Science Central platform too for any organization or person to use, Inkspot,” said Watson.

The e-Science Central cloud will eventually support the Ambient Kitchen and a number other projects under the SIDE initiative, which has a budget of £12 million over five years from the UK Research Councils.

The SIDE project is dealing with a number of sensitive social issues with cloud computing supporting its infrastructure. With clouds being so prolific, the question of data security arises. Watson thinks that a tradeoff has to be made between ease-of-access to data and confidentiality.

”As an administrator on e-Science Central you can make information public or private; read only or write only. If you need to send sensitive data to a public cloud for computationally-intensive processing, for example to Amazon EC2, this can first be anonymized so no personal information is included. All a user has is a username and password; but, if you add additional security like an X-509 certificate that is used for grid computing, then you reduce usability to increase the security, which makes it less practical.”

Your rating: None Average: 4 (2 votes)

Comments

Post new comment

By submitting this form, you accept the Mollom privacy policy.