| Acceleration | 0-100km/h: 2,0s |
|---|---|
| Power Output | 4x28kW |
| Weight | 226kg |
PWd3.19
| Monocoque | CFRP one-piece |
|---|---|
| Aerodynamics | Side and rear wing, skid plate and splitter |
| Motor | 4 wheel hub motors |
|---|---|
| Top performance | per Motor: 28 kW |
| Gearbox | 1 ½ stage planetary gear |
| Battery container | Central battery container with 7.5 kWh |
|---|---|
| Cells | Samsung INR 18650 25 R |
| Maximum voltage | 580V |
| Control unit | ETAS |
|---|---|
| Driving dynamics control systems | Torque vectoring, recuperation, traction control, power limitation |
| Telemetry | Live Telemetry via 2.4 GHz WLAN |
| Dashboard | Self-developed dashboard |
| Wheelbase | 1540mm |
|---|---|
| Front track width | 1150mm |
| Rear track width | 1128mm |
| Tires | 7,5″x10″ Hoosier Slicks |
| Rims | 10″ CFRP rims with aluminum star |
| A-Arm | Double A-Arm |
| Dampers | Pushrod actuated damper |
| Cameras | 2x Intel D435, 1x Bosch SVC2 |
|---|---|
| Sensors | Kistler Correvit SFII, Bosch MM5.10-R acceleration sensor, VectorNav VN-300 INS |
| Processing Units | 2x NVIDIA Jetson TX2, ETAS ES910 |
| Environment recognition | Object recognition via a neural network |
|---|---|
| Mapping / Localization | In-house developed C ++ libraries with a focus on storage space optimization |
| Test environment | Hardware in the loop test bench |
| Toolchain | C++, Python, Matlab, IPG Carmaker |
| Trajectory planning | Model-based off-policy reinforcement learning to optimize lap times |
| Training automation | In-house developed software for data set management, automatic hyperparameter optimization and management of machines in the training cluster |
