Unmanned Systems Technology 020 | Alpha 800 I Additive Manufacturing focus I USVs insight I Pegasus GE70 I GuardBot I AUVSI Xponential 2018 show report I Solar Power focus I CUAV Expo Europe 2018 show report

18 “C ompletely by accident” is how Chris Hoyle, 50, says he got into the simulation of unmanned vehicles and the environments they operate in. Now technical director of (and major shareholder in) rFpro – whose customers include Formula One and World Endurance Championship teams, as well as car manufacturers, unmanned vehicle systems developers and AI specialists – he took a very unconventional route into the industry. A computer programmer since the age of 12 and a lifelong car and motorcycle enthusiast, he started creating games while still at school, in the early 1980s. Fast forward to the early 2000s and he was still programming for fun. Combining coding with his passion for fast things on two wheels and four, and the Caterham Seven in particular, he decided to create a virtual model of the sportscar he calls his idea of perfection, along with a shrink-wrapped engine and driver. “I wanted to simulate a Caterham Seven, but I didn’t want it to be superficial like a computer game,” he says. “I wanted it to be as perfect as possible, so I ended up creating a Pareto optimal inverse kinematic solver.” An inverse kinematic solver is a piece of software that recreates the dynamics of a real-world object from other data, such as video of a car in motion, measurements of accelerations and so on. If something is ‘Pareto optimal’ it means the solutions it produces cannot be improved in any one area without degrading the result in another. In 2006, he made this program available for free on the internet, and it was picked up by a couple of commercial users. One of those was a team building a Le Mans car, the other was a Formula One team. “The Formula One team thought I might be a useful chap, so they called me in, and over the course of a couple of hours with the head of r&d we effectively designed their driving simulation strategy,” he says. “So my hobby led first of all to a bit of part-time fun, then within three months it was full-time fun, and by the end of 2007 I thought, ‘Well, this is actually the most fun I’ve ever had in my life’ and it has now turned into a business.” The technical director of rFpro gives Peter Donaldson an insight into what drives his enthusiasm for vehicle simulation Virtual passion June/July 2018 | Unmanned Systems Technology Using virtual worlds to teach AI software to drive, rFpro creates training data sets like this simulation of a challenging junction with road markings, traffic, signage, pedestrians and vegetation (Images courtesy of rFpro)