Sequential quadratic programming-based fast path planning algorithm subject to no-fly zone constraints

Path planning plays an important role in aircraft guided systems. Multiple no-fly zones in the flight area make path planning a constrained nonlinear optimization problem. It is necessary to obtain a feasible optimal solution in real time. In this article, the flight path is specified to be composed...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Engineering optimization Ročník 48; číslo 8; s. 1401 - 1418
Hlavní autoři: Liu, Wei, Ma, Shunjian, Sun, Mingwei, Yi, Haidong, Wang, Zenghui, Chen, Zengqiang
Médium: Journal Article
Jazyk:angličtina
Vydáno: Abingdon Taylor & Francis 02.08.2016
Taylor & Francis Ltd
Témata:
Air traffic control
Algorithms
Circularity
intersection point
Mathematical analysis
no-fly zone
Nonlinear programming
Nonlinearity
Optimization
Path planning
polygon
Quadratic programming
Segments
SNOPT
ISSN:0305-215X, 1029-0273
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Path planning plays an important role in aircraft guided systems. Multiple no-fly zones in the flight area make path planning a constrained nonlinear optimization problem. It is necessary to obtain a feasible optimal solution in real time. In this article, the flight path is specified to be composed of alternate line segments and circular arcs, in order to reformulate the problem into a static optimization one in terms of the waypoints. For the commonly used circular and polygonal no-fly zones, geometric conditions are established to determine whether or not the path intersects with them, and these can be readily programmed. Then, the original problem is transformed into a form that can be solved by the sequential quadratic programming method. The solution can be obtained quickly using the Sparse Nonlinear OPTimizer (SNOPT) package. Mathematical simulations are used to verify the effectiveness and rapidity of the proposed algorithm.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:0305-215X
1029-0273
DOI:10.1080/0305215X.2015.1111085