With the high number of accidents affecting cyclists in South Africa, a new invention by a Stellenbosch-based technology company called iKubu, which alerts bike riders of what’s happening around them, is likely to be a welcome development.
iKubu, which specialises in radar and computer vision technology, has developed a bike-mounted radar called Backtracker for cyclists that alerts them of approaching vehicles. It’s already attracting plenty of attention among the cycling fraternity worldwide.
iKubu was founded in 2006 by business partners Franz Struwig and Denho Geldenhuys, who had studied engineering together at the University of Pretoria. Two years later, they headhunted another fellow Tukkies engineering student, Nolan van Heerden, who was the top performer in their class and is iKubu’s “radar guru”.
“When iKubu started, it did not have a technical objective,” says Struwig, who serves as iKubu’s MD. “It was counter to corporate culture. We wanted to create a space where engineers could express themselves. We believe technology should be used to supercharge our sensory abilities and we see computer vision and radar as a way of achieving this.”
iKubu develops products that typically contain plenty of electronics and complex algorithms that require intensive software design.
The idea for Backtracker was born in 2010 when the team wanted a technology solution to help reduce the risk associated with cycling in traffic. As a group of avid cyclists, the team had a vested interest in road safety, says Struwig. Research showed — perhaps not surprisingly — that rear collisions with bicycles is one of the biggest causes of cyclist deaths.
The team built a prototype bike-mounted radar within a month and sought market feedback. “What we realised was that this was just sixth sense [using the radar]; it becomes quite intuitive once you start using it.”
iKubu spent a year trying to secure South African and international investors for Backtracker. Instead, they decided to take the product to market on their own through a crowdfunding campaign.
With 20 days left of the crowdfunding initiative, iKubu has already raked in about US$100 000 of a goal of $194 500. Backers should get the product in their hands by December. Early funders will get it for $179. Thereafter, it will cost $199.
How does it work? Backtracker is a bike-mounted radar kit that consists of two parts. The first part, the radar, is mounted on the back of the bike, while the second, a notification unit, is fitted to the handlebars.
The rear-facing radar doubles as a warning light for motorists. The radar communicates with the handlebar indicator unit. The indicator then displays the proximity of vehicles behind the cyclist using a series of LEDs, indicating their movement as they near his or her position.
The technology only detects traffic behind the cyclist and won’t pick up vehicles driving in the opposite direction.
Backtracker is able to interpret the speed and acceleration of vehicles approaching from the rear at a distance of up to 140m.
“The radar technology is like a sixth sense for cyclists, especially in low visibility conditions such as rain and fog and works well in early morning and late afternoons,” Struwig says.
He also says that the radar is not affected by other cyclists — useful for cyclists riding in groups.
iKubu’s engineers have worked on a number of other projects, including the development of a 3D scanning system for Qantas that allows the Australian airline to automate the baggage check-in process.
The technology has already been deployed across Australia, at London’s Heathrow airport, and at Charles de Gaulle in Paris.
iKubu’s computer vision technology creates “point clouds” and 3D imaging of each bag put on an automated bag-drop conveyor belt.
“Our algorithms tell their systems if a bag should be accepted or not and dictate what kind of feedback it should give to the user,” says Struwig.
iKubu is also involved in a project to automate the counting of Africa’s elephant populations. It does this by mounting infrared and RGB cameras on a plane and letting algorithms classify the wildlife it detects, effectively sorting rhinos from elephants.
The company, called SkyReach, was founded by Paul Maritz, the Zimbabwean-born and South African-educated computer scientist who has held top positions at Intel, Microsoft and VMware. The project also enjoys funding from Paul Allen, the billionaire who co-founded Microsoft with Bill Gates.. — © 2014 NewsCentral Media