Aircraft trajectory planning with collision avoidance using mixed integer linear programming

Richards, Arthur; How, Jonathan P.

doi:10.1109/acc.2002.1023918

Cited by 582 publications

(365 citation statements)

References 7 publications

Supporting

Mentioning

350

Contrasting

Unclassified

Order By: Relevance

“…Based on the assumption of the Markov properties, the robot next state depends on the current state s ∈ S 0 and the current decision a ∈ A 0 , so Q-learning does not have to establish an environmental model. It establishes an iterative optimization strategy directly based on a rewardpunishment mechanism (Richard and Andrew 1998 In this paper, RST also needs to create state lists of the environment. If all sub-state Cartesian product is to be stored, a large number of meaningless and redundant states existing in learning samples must lead to a low real-time performance in practice.…”

Section: Comparison With Other Methodsmentioning

confidence: 99%

Knowledge-driven path planning for mobile robots: relative state tree

Yang

Cheng

et al. 2014

Soft Comput

View full text Add to dashboard Cite

Path planning is important in the field of mobile robot. However, traditional path planning techniques optimize the navigation path solely based on the models of the robot and the environments. Owing to the time-varying environment, the robot is expected to launch the replanning procedure in real-time continuously. It is slow and wastes computing resources for repeated decisions. In this study, a new perspective is adopted which utilizes a knowledge-driven approach for path planning. The concept of relative state tree is proposed to develop an incremental learning method based on a path planning knowledge base. The knowledge library, which stores a collection of the mappings from environmental information to robot decisions, can be established by offline or online learnings. As the robot plans online, its movement is guided by the optimal decision that is retrieved from the library based on the information which matches mostly the current environment. A large number of simulations are executed to verify the proposed method. When Communicated by G. Acampora.

show abstract

Section: Comparison With Other Methodsmentioning

confidence: 99%

Knowledge-driven path planning for mobile robots: relative state tree

Yang

Cheng

et al. 2014

Soft Comput

View full text Add to dashboard Cite

show abstract

“…However, one simple way to represent these constraints in a linear formal way is shown in Eq. (16). The profile of these constraints, a cube, is shown in Fig.…”

Section: Constraints On Feasible Regionmentioning

confidence: 99%

“…Based on the relative velocity obstacle method, 13 linear programming (LP), nonlinear programming (NLP), mixed integer linear programming (MILP), and other convex optimization methods have been employed for path planning problem. [14][15][16] For example, Wang, et al, 14 converted the 2D path planning of an unmanned underwater vehicle to constrained optimization or semi-infinite constrained optimization problem. Zu et al 15 discussed the path planning in two dimensions and an LP method was proposed for the problem of dynamic target pursuit and obstacle avoidance (TPOA).…”

Section: Introductionmentioning

confidence: 99%

Three-dimensional path planning for unmanned aerial vehicle based on linear programming

2011

View full text Add to dashboard Cite

SUMMARYIn this paper, an approach based on linear programming (LP) is proposed for path planning in three-dimensional space, in which an aerial vehicle is requested to pursue a target while avoiding static or dynamic obstacles. This problem is very meaningful for many aerial robots, such as unmanned aerial vehicles. First, the tasks of target-pursuit and obstacle-avoidance are modelled with linear constraints in relative coordination according to LP formulation. Then, two weighted cost functions, representing the optimal velocity resolution, are integrated into the final objective function. This resolution, defined to achieve the optimal velocity, deals with the optimization of a pair of orthogonal vectors. Some constraints, such as boundaries of the vehicle velocity, acceleration, sensor range, and flying height, are considered in this method. A number of simulations, under static and dynamic environments, are carried out to validate the performance of generating optimal trajectory in real time. Compared with ant colony optimization algorithm and genetic algorithm, our method has less parameters to tune and can achieve better performance in real-time application.

show abstract

“…Trajectory generation for autonomous vehicles was considered in Schouwenaars et al (2001), while a robotic ball game was the application of Earl and D'Andrea (2002). Since then MILP has been used extensively for path planning problems, in particular for UAVs, both for single and multi-vehicle systems, see Richards and How (2002), Ma and Miller (2006), Shengxiang and Hailong (2008), Kuwata and How (2011), Grøtli and Johansen (2012b) and Grøtli and Johansen (2012c). In Schouwenaars et al (2006) connectivity constrained trajectory planning for autonomous helicopters through cluttered environments was studied.…”

Section: Introductionmentioning

confidence: 99%

“…Among the first to consider multi-vehicle path planning using MILP, were Schouwenaars et al (2001), Earl and D'Andrea (2002) and Richards and How (2002). Trajectory generation for autonomous vehicles was considered in Schouwenaars et al (2001), while a robotic ball game was the application of Earl and D'Andrea (2002).…”

Section: Introductionmentioning

confidence: 99%

Motion- and Communication-Planning of Unmanned Aerial Vehicles in Delay Tolerant Network using Mixed-Integer Linear Programming

Grøtli

Johansen

2016

MIC

View full text Add to dashboard Cite

Large amounts of data are typically generated in applications such as surveillance of power lines and railways, inspection of gas pipes, and security surveillance. In the latter application it is a necessity that the data is transmitted to the control centre "on-the-fly" for analysis. Also missions related to other applications would greatly benefit from near real-time analysis and operator interaction based on captured data. This is the motivation behind this paper on coarse offline motion-and communication-planning for cooperating Unmanned Aerial Vehicles (UAVs). A Mixed-Integer Linear Programming (MILP) problem is defined in order to solve the surveillance mission. To efficiently transmit the data back to the base station the vehicles are allowed to store data for later transmission and transmit via other vehicles, in addition to direct transmission.The paths obtained by solving the optimization problem are analyzed using a realistic radio propagation path loss simulator. If the radio propagation path loss exceeds the maximum design criterion the optimization problem is solved again with a stricter communication constraint, and the procedure is continued in an iterative manner until the criterion is met. The proposed algorithm is supported by simulations showing the resulting paths and communication topologies for different choices of delay tolerance.

show abstract

Aircraft trajectory planning with collision avoidance using mixed integer linear programming

Cited by 582 publications

References 7 publications

Knowledge-driven path planning for mobile robots: relative state tree

Knowledge-driven path planning for mobile robots: relative state tree

Three-dimensional path planning for unmanned aerial vehicle based on linear programming

Motion- and Communication-Planning of Unmanned Aerial Vehicles in Delay Tolerant Network using Mixed-Integer Linear Programming

Contact Info

Product

Resources

About