A novel algorithm for collision avoidance in commercial aircraft using neural networks and non-linear programming

Christodoulou, Manolis A.; Kontogeorgou, Chrysa

doi:10.1109/med.2008.4601988

Cited by 4 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remark: Constant speed vectors assumption is not restricting, since conflict detection conditions (6) can be evaluated with properly high update rate 1/T S , according to the Nyquist-Shannon sampling theorem. Consequently, reference signals to the autopilot are generated with sampling time T S .…”

Section: Fig 4 Collision Geometry Between a Point Object A And A Cymentioning

confidence: 99%

“…In this case, conflict detection conditions (6) can be evaluated with properly high update rate 1/T S , according to the Nyquist-Shannon sampling theorem and, consequently, reference signals to the autopilot are generated with sampling time T S . On the other hand, if the non-cooperative intruder does not move along a straight trajectory or, in the worst case, if it is a pursuer, there is no guarantee that A/C A will not enter the safety cylinder surrounding A/C B .…”

Section: Aircraft Envelope Limitations and Dynamic Constraintsmentioning

confidence: 99%

“…Conditions (6) also offer an analytical criterion to find initial positions and speed vectors of the two aircrafts which cause initial relative velocity vector to enter in the safety bubble. By considering only these conflict scenarios, the amount of collision geometries to be analyzed is dramatically reduced.…”

Section: Numerical Validationmentioning

confidence: 99%

“…The main disadvantage of this approach is that the algorithm, although considering a simplified planar kinematic aircraft model having only three states, takes several minutes to compute the safe region. In [6] the authors presented a neural network and non-linear programming based algorithm to solve a collision avoidance problem for commercial aircrafts. The main disadvantage of the method proposed in this work consists in considering always and only a spherical safety bubble, as usual, without examining any possible alternative.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A novel 3D analytical algorithm for autonomous collision avoidance considering cylindrical safety bubble

Luongo

Corraro

Ciniglio

et al. 2010

2010 IEEE Aerospace Conference

View full text Add to dashboard Cite

This paper presents an innovative 3D analytical algorithm for the resolution of the pair-wise noncooperative collision avoidance problem between aircrafts. The proposed algorithm addresses the above described problem by using an innovative approach, based on the consideration of a cylindrical safety bubble, and it is able to obtain an optimal three-dimensional analytical solution for this problem. This novel approach allows different minimum separations on the vertical and horizontal planes with respect to the nominal trajectory to be achieved, so minimizing the impact of the collision avoidance maneuver on surrounding traffic. Moreover, the algorithm has the very interesting feature that it does not require the solution of any non deterministic and/or iterative problem, resulting suitable for real-time applications. This is due to the capability of the algorithm to find a closed form solution for the kinematic optimization problem here considered. The solution of the collision avoidance problem requires the simultaneous change of all control variables (speed module, track and slope angles), aiming to assure the required safety level and, at the same time, to minimize aircraft deviation from the nominal trajectory. This system is mainly developed for unmanned aircraft vehicles, where high levels of autonomy (i.e. the avoidance maneuver is autonomously executed by a standard autopilot) are required, but it can also be used, as aid to pilots, in manned commercial aircrafts. The effectiveness of the algorithm is evaluated by means of numerical simulations, where suitable conflict scenarios, taking into account aircraft dynamics and on-board sensors errors and limitations, are considered. Scenarios where both aircrafts are equipped with the proposed collision avoidance algorithm or where both aircrafts are subjected to Visual Flight Rules are also considered. 1 2

show abstract

Section: Fig 4 Collision Geometry Between a Point Object A And A Cymentioning

confidence: 99%

Section: Aircraft Envelope Limitations and Dynamic Constraintsmentioning

confidence: 99%

Section: Numerical Validationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations