"Range anxiety" is not just affecting electric vehicle drivers on the road—it is also a significant hurdle for Formula E teams on the track. Highly accurate energy consumption prediction using knowledge not just of race circuit layout but also of track surface detail is vital.
The all-electric Formula E race series, supported by a growing list of OEMs including Audi, BMW, Jaguar, Mercedes and Renault, requires every team to find ways of extracting, and making best use of, all the potential energy from the battery pack by the time the car crosses the finishing line. This is to enable a driver to use maximum power in appropriate situations, otherwise the car could have gone faster at some point in the race.
“Optimizing the energy consumption by adjusting the calibration before the race relies on having an accurate circuit model," explained Chris Hoyle, Technical Director at rFpro, a U.K.- based simulation software company. "But with many circuits based in parks or city centers, the exact geometry of every corner may not be known until the barriers are erected.”
There are further complications. Apart from variations in circuit layout, accurate prediction of energy usage is made more difficult by the irregularity of many track surfaces, Hoyle noted. Bumps, road repairs, pot holes or drain covers could result in loss of control or reduced grip, often forcing drivers to take a completely different line to the shortest or theoretically quickest racing ideal.
Conventional circuit mapping, such as using Google Earth data, fails to give a true representation of the vehicle path around the full lap, he said. Knowledge of a circuit’s full details is a must; in the precise and challenging world of E racing, a car's electric energy consumption can make the difference between a podium place and a DNF.
To cope with this, rFpro has developed TerrainServer, a simulation package designed to accurately reproduce literally every bump, curb, ripple and degree of camber of a track, feeding high-bandwidth, high-fidelity, cleaned-LiDAR-point cloud data for each tire contact patch into the vehicle model in real time at up to 5kHz.
As Formula E racing gains popularity, the category is gradually assuming F1 levels of sophistication. Hoyle's company is now supporting the series, complementing their established driver-in-the-loop (DIL) simulator software solutions for Formula One, NASCAR and World Endurance Championship teams.
The company now has most of the established Formula E circuits scanned and modeled to the same levels of precision as the world’s F1 tracks. Its improved modeling accuracy helps the cars make full use of all available energy, to run faster for longer. With regulations placing limitations on Formula E chassis design, rFpro has found that teams are increasingly concentrating simulation time on improving the accuracy of their energy consumption predictions—and developing their complex control systems.
Overcoming the first challenge—circuit layout—is achieved by scanning after the barriers have been erected and updating the data in the circuit model which is used by the simulator. The second issue, accounting for road surface features, has only been made possible by two fundamental developments in DIL performance: faster response and improved surface data capture.
“To ensure that a driver behaves exactly the same in the simulator as on the track, the experience must be totally convincing. This means all the cues—visual, aural, haptic—must arrive on schedule in real time,” Hoyle explained. He noted that some older-generation simulators had up to .25 s latency (delay) which meant they were limited in their capabilities and too slow for chassis dynamics work.
The rFpro software provides video signals ten times faster and audio signals 20 times faster, which overcomes this limitation, he claimed.
Having provided a convincing simulation environment in which the vehicle responds in real time, the further key ingredient is always the accurate road surface model. It provides the correct inputs to the vehicle’s tires—the task of TerrainServer.
“The pressure on teams to extract the maximum possible energy from their car’s battery pack by the time it crosses the finishing line is likely to intensify if proposed 2018-2019 rule changes go ahead,” Hoyle added. These will see the end of drivers switching to a second car during the race.
Having only one car, with one battery pack as an energy source, will place even more emphasis on extracting the maximum without exhausting their energy supply prematurely. "Highly accurate simulation can help to achieve that,” he said.