Stochastic Model Predictive Control for Collision Avoidance and Landing of Aircraft

Both the air traffic demand and the use of drones have continued to expand recently. As a result, the number of collision accidents between aircraft and drone has increased. We propose a stochastic model predictive control (SMPC) system for collision avoidance and landing which takes uncertain information of wind and obstacle positions into consideration. We carried out vertical and lateral simulations with static and moving obstacles using linear aircraft model. The simulation result showed that the controller was able to maintain the reference trajectory. Our proposed SMPC system for collision avoidance also proved to be effective for avoiding static obstacles with constant linear motion. Further improvements are needed to avoid obstacles with more complex, random movement.