Shell Eco-Marathon Simulation

Versions 3 through 5 expanded the simulation from a parameter study tool into a closed-loop vehicle model capable of autonomous speed control and realistic engine behavior. These versions introduced automated driver control, real engine performance data, and physically realistic torque modeling. Together these improvements established the core simulation architecture used in later versions.

Key Enhancements

Driver Strategy (V3)

Closed-loop driver model implemented
PID speed controller added
Automatic throttle control (0–1 range)
Target speed tracking with velocity feedback integration

Engine Modeling (V4)

Dynamometer data imported into MATLAB
Horsepower converted to torque; torque vs RPM lookup tables generated
Realistic engine operating limits enforced
Gear ratio linked to engine speed

Driver Strategy Refinement (V5)

Speed bands added around target velocity
Hysteresis thresholds implemented to reduce control oscillation
Extended simulation time support and improved parameter organization

MATLAB & Simulink Scripts

shell_eco_paramsV4.m — Parameter configuration
build_engine_mapV4.m — Dyno data processing
engine_mapV4.mat — Torque lookup table
analyzeV4.m — Automated analysis

View Source Code on GitHub

Simulation Gallery

Limitations

Vehicle speed remained nearly constant — throttle rarely reached zero
Engine operated continuously, never truly coasting
Model behaved like cruise control rather than true pulse-and-coast driving
Did not represent the actual Shell Eco-Marathon driving strategy

Future Work (V6)

Conclusion

Versions 3 through 5 transformed the simulation into a realistic closed-loop vehicle model with experimental engine data integration. These developments provided the foundation for the Version 6 pulse-and-coast driver model.

Shell Eco-Marathon Vehicle Simulation — V3–V5

Overview