Perception-Aware Motion Control of Multiple Aerial Vehicle Transportation Systems

How to ensure that multiple aerial vehicle transportation systems can accurately track a target despite the dynamical constraints? In a general Unmanned Aerial System, the system can identify the target by sensors to adjust its flight status. However, for Multiple Aerial Vehicle Transportation Systems, if each unmanned aerial vehicle has sensors with different fields of view on board, it creates conflicts in the system and does not enable proper sensing. Moreover, the added perceptual requirements in the system will also cause conflicts with the corresponding flight dynamics constraints and will have some impact on the flight path. In this regard, we add a perception model for Multiple Aerial Vehicle Transportation Systems in Model Predictive Control to solve these conflicting problems. In this paper, we first add a perceptual constraint model of the transportation system with the camera as the sensor to the previously established dynamics simulation model. And the model is applied to multiple UAVs so that multiple cameras can aim at the same target for tracking. Then, we can solve feasible UAV motion states by incorporating the MPC of the perception model. In this way, we provide a feasible method for multi-sensor information fusion for Multiple Aerial Vehicle Transportation Systems. Finally, the simulation results demonstrate the effectiveness of the method.