To ensure real-time conflict detection and resolution, the whole flight conflict
detection and resolution time of the two methods is as shown in Fig. 10.
Flight conflicts are detected by aircraft in the 7th and 18th cycles. The detection
time is shorter than 0.3 s on Python platform. In post-conflict detection, the violent
search method takes longer time to extricate, which does not meet the real-time
requirements, while the dynamic programming method takes shorter time. It can ensure
a certain real-time performance on other computing platforms for the dynamic programming method.
5 Conclusions
In order to ensure air traffic safety and reduce the situation of controllers in the complex
conflict conditions of multiple aircraft, this paper proposes a method based on digital
grid to detect and release flight conflicts. This method improves the automation performance of the air traffic control system. The main advantages are: (1) The prediction
range is larger than the recommended decision range of the aircraft TCAS system,
auxiliary air traffic controllers make command decisions on multiple aircraft; (2) Provides conflict resolution optimal solutions for TCAS system failures or unreliable
conditions. The conflict detection and liberation in this paper are based on the digital
grid model of the airspace which sacrifices part of the space and greatly simplifies the
calculation of the safety interval. The motion model of the aircraft is fully considered to
obtain the reachable area for collision detection, and the dynamic planning method is
used to obtain all the non-conflicting maneuver state combinations of each aircraft.
Finally, the optimal conflict resolution decision is obtained according to the performance index. At the same time, the current liberation algorithm still has some shortcomings, and the related problems of the simulation algorithm data structure and
optimal parameters need to be further optimized.
(a) Dynamic programming method
(b) Violent search methods
Fig. 10. Detection release time
478
Y. Yi et al.
detection and resolution time of the two methods is as shown in Fig. 10.
Flight conflicts are detected by aircraft in the 7th and 18th cycles. The detection
time is shorter than 0.3 s on Python platform. In post-conflict detection, the violent
search method takes longer time to extricate, which does not meet the real-time
requirements, while the dynamic programming method takes shorter time. It can ensure
a certain real-time performance on other computing platforms for the dynamic programming method.
5 Conclusions
In order to ensure air traffic safety and reduce the situation of controllers in the complex
conflict conditions of multiple aircraft, this paper proposes a method based on digital
grid to detect and release flight conflicts. This method improves the automation performance of the air traffic control system. The main advantages are: (1) The prediction
range is larger than the recommended decision range of the aircraft TCAS system,
auxiliary air traffic controllers make command decisions on multiple aircraft; (2) Provides conflict resolution optimal solutions for TCAS system failures or unreliable
conditions. The conflict detection and liberation in this paper are based on the digital
grid model of the airspace which sacrifices part of the space and greatly simplifies the
calculation of the safety interval. The motion model of the aircraft is fully considered to
obtain the reachable area for collision detection, and the dynamic planning method is
used to obtain all the non-conflicting maneuver state combinations of each aircraft.
Finally, the optimal conflict resolution decision is obtained according to the performance index. At the same time, the current liberation algorithm still has some shortcomings, and the related problems of the simulation algorithm data structure and
optimal parameters need to be further optimized.
(a) Dynamic programming method
(b) Violent search methods
Fig. 10. Detection release time
478
Y. Yi et al.