Projects

Team: Baymax

Members:

• Nathanaël Bullier
• Tom Meltzer
• Damien de Mijolla
• Krishna Jadeja

Project:

A voice-controlled, raspberry pi helper robot that will provide emergency help to; people with reduced mobility or specific health needs; elderly people living alone and the visually impaired. The helper robot utilizes; a raspberry pi which interfaces the robot with the amazon echo dot (used for voice control), a camera and an ultra-sound sensor for interacting with the environment. Google's Cloud Vision API is accessed through the raspberry pi in order to provide object identification for objects in the camera's field of vision. All connections are handled via ssh using the raspberry pi zero w's on-board Wi-Fi.

GitHub page: https://github.com/TomMelt/baymax

Team: 2Fast2Furious

Members:

• Michael Staddon
• Josh Tingey
• Danial Dervovic
• Jack Poulton

Project:

Transport for London (TfL)’s rent-a-bike scheme is used by millions of commuters a year. Often users find that docking stations are empty when need a bike and full when they want to drop off their bike. To solve this, TfL have a fleet of 30 vans that drive around redistributing bikes; their strategy at the moment is reactive, they are called up when a docking station is close to full or empty. Our project was to use 5 years of rental data (over 2GB!) to improve redistribution strategy to be more predictive of demand. This involved correlating bike usage with variables such as the weather, time of day and year. We also built a simulation of London’s cycle network to test these new strategies.

Team: Happy Haptics

Members:

• Sofia Qvarfort
• Omar Jahangir
• Kristof Farkas-Pall
• Luciana Barros Henaut

Project:

Navigate a virtual maze using the Ultrahaptics board as input and output.

Team: In Our Face

Members:

• Muhammad Javed Akram
• Simone Famiani
• Joshua Lockhart
• Momchil Konstantinov

Project:

Using a Raspberry Pi with a camera and biometric sensors to evaluate someones mood.

Team: Feel Your Way

Members:

• Patrick Bolton
• James Legg
• Matthew Flinders
• David Wise

Project:

Using a Leapmotion controller to control a Raspberry Pi buggy, and an Ultrahaptics board to provide feedback on obstacles.

Team: Home ID

Members:

• Rhys Evans
• Yuval Cohen
• Alice Morris
• David Yallup

Project:

Building a portable facial recognition system.

Team: Plant Doctors

Members:

• Matt Ellis
• Kevin Mulder
• Ricky Nathvani
• Amy Lai

Project:

We use a hardware set up connected to a raspberry pi to measure the ambient conditions (temperature, humidity, soil moisture, light exposure, a photo of the plant) in a given indoor environment. We then pass this data on to an algorithm which combined photo sensor data to determine how well suited your environment is for your plant, (including a machine learning algorithm that can determine the health of the plant's leaves from a photo). The algorithm also determines, from a look-up in a customised database, the plant most ideally suited to the measured environment. The process was developed into a one-click operation web app that also recommends nearby plant shops in response to queries.

Team: VR Xposure

Members:

• Henry Banks
• Sebastian Bending
• Amal Vaidya
• Caishan Quek

Project:

Designed virtual reality environments to trigger phobias with the aim of helping people overcome them.