We apply vehicle and trajectory co-design to quadcopters flying through gates and around obstacles. The fastest feasible path depends on propulsion, mass, and rotor geometry, so a trajectory optimized for one vehicle need not suit another.
Designing the vehicle for the course
The study combines differentiable flight and propulsion models with direct collocation and geometric path constraints. Comparisons across gate courses illustrate how the preferred vehicle configuration changes with the maneuvering task, linking hardware choices to the flight profile they enable.
Study: Soliman Villapando, Simultaneous Vehicle and Trajectory Optimization of Unmanned Aerial Vehicles in Constrained Environments, master's thesis presentation.
Optimization problem
| Case | Quadrotor co-design for gate and obstacle courses |
|---|---|
| Objective | Minimize course completion time. |
| Design variables | 14 vehicle variables: motor mass, battery mass, arm length, rotor radius, five blade-twist values, and five blade-chord values; trajectory states, controls, and final time. |
| Constraints | Collocation dynamics and initial/final conditions; rotor speed between 0 and 30,000 RPM; motor torque, motor power, and battery power limits; gate-crossing and obstacle-avoidance conditions. |
| Models and conditions | Rigid-body flight dynamics, quasi-steady rotor inflow, and electromechanical scaling models, with aerodynamic loads evaluated in BladeAD and derivatives computed in CSDL. |
| Representative source | Soliman S. Villapando, Simultaneous Vehicle and Trajectory Optimization of Unmanned Aerial Vehicles in Constrained Environments, master’s thesis presentation. |
Research connections