Conference Paper

Distributed Control for Multiple UAVs with Strongly Coupled Tasks

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

This paper addresses the development of algorithms for the dynamic assignment of multiple, strongly coupled tasks in cooperative decision and control of a distributed UAV team. A symmetric vehicle-target scenario is considered with extensive task precedence and timing constraints. Moreover, strong coupling of task assignment with task planning can lead to in- feasible solutions. A suite of algorithms to address the scenario is presented. These algorithms are analyzed with respect to feasibility, optimality, strength and type of coupling that can be accommodated, required communications, and degree of decentralization. Algorithms discussed include: mixed integer linear programming, auction based optimization, constraint satisfaction based feasibility enforcement, network ßow, and hybrid techniques with heuristics. Analysis has revealed that, for strongly coupled tasks, decentralized solutions are not necessarily better than centralized solutions, unless communication costs are low.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The network flow model has been further extensively studied by Schumacher [2, 3, 4] for wide area search munitions with variable path lengths, assignment with timing constraints and path planning. Chandler et al. [5, 6] , explore various other techniques like iterative network flow, auctions, linear programming , and mixed integer linear programming, for multi-UAV task allocation. The effect of communication delays on the task allocation using the iterative network flow model is dealt with in Mitchell [7]. ...
... These authors also address the problem of obstacle avoidance for the UAV. Jin et al.[10] propose a probabilistic task allocation scheme for the scenario presented in [5, 6]. In most of the algorithms presented in the papers cited above, global communication between agents is assumed although the agents themselves have limited sensor range. ...
Conference Paper
Full-text available
A multiple UAV search and attack mission in a battlefield involves allocating UAVs to different target tasks efficiently. This task allocation becomes difficult when there is no communication among the UAVs and the UAVs sensors have limited range to detect the targets and neighbouring UAVs, and assess target status. In this paper, we propose a team theoretic approach to efficiently allocate UAVs to the targets with the constraint that UAVs do not communicate among themselves and have limited sensor range. We study the performance of team theoretic approach for task allocation on a battle field scenario. The performance obtained through team theory is compared with two other methods, namely, limited sensor range but with communication among all the UAVs, and greedy strategy with limited sensor range and no communication. It is found that the team theoretic strategy performs the best even though it assumes limited sensor range and no communication.
... The network flow model has been further extensively studied by Schumacher [2] [3] [4] for wide area search munitions with variable path lengths, assignment with timing constraints and path planning. Chandler et al. [5] [6], explore various other techniques like iterative network flow, auctions, linear programming, and mixed integer linear programming, for multi-UAV task allocation. The effect of communication delays on the task allocation using the iterative network flow model is dealt with in Mitchell [7]. ...
... These authors also address the problem of obstacle avoidance. Jin et al. [10] propose a probabilistic task allocation scheme for the scenario presented in [5] [6]. In most of the algorithms presented in the papers cited above, global communication between agents is assumed. ...
Conference Paper
Full-text available
A multiple UAV search and attack mission involves searching for targets in an unknown region, followed by attack on the detected targets. An effective mission involves assigning the tasks to UAVs efficiently. Task allocation becomes difficult when the UAVs have limited range to detect the targets and neighbouring UAVs. The UAVs are also subject to limited communication range. With these constraints, allocating tasks efficiently to UAVs become difficult. In this paper, we propose a negotiation scheme that efficiently allocates tasks for multiple UAVs. The performance of the negotiation based task allocation is studied on a battle field scenario. We study the effect of sensor ranges on the task allocation and compare the results with that of greedy strategy and a variation of the negotiation mechanism. The results show that the negotiation based task allocation mechanism performs far better than the greedy strategy. Negotiation with target information exchange between neighbours performs better than without target information exchange, when the sensor radius is low.
... Over the past several years Miniature Aerial Vehicles (MAVs) have received considerable attention for many useful applications ranging from battlefield surveillance , smart munitions and real time bomb damage assessment to forest -fire reconnaissance, surveys of natural disaster areas and inexpensive traffic and accident monitoring [1][2][3][4][5][6][7][8][9][10]. In this, the ability of the MAV to fly in constrained environments is an important requirement for both civil and military applications. ...
Article
Full-text available
This paper uses the vector field approach to represent both the desired ground track heading and the vertical flight path angle to navigate unmanned aerial vehicles onto the desired path. The path following is realized in 3-dimensions, thus allowing the aircraft to navigate longitudinally as well as laterally, unlike the current approaches which require the aircraft to maintain a constant altitude. This results in an increase in the maneuverability (like overflying a hill) of the vehicle allowing it to navigate over various terrains and terrain following. Methods for following straight lines in 3-dimensional space are demonstrated through numerical simulations.
... For these conditions, we propose the algorithm for pre-flight route to the target area planning, based on Delaunay triangulation [25] (Fig. 1), which provides building-up the optimal UFV routes, taking into account the above listed requirements, by resolving the mathematical programming problem, using criterion: ...
Article
Full-text available
In article presents algorithmic for a reconnaissance and combat unmanned flying vehicle (UFV) group pre-flight planning. The algorithms, which we elaborated, includes: algorithm, involved for mission target decision, membership selection. Discussing algorithm is based on analytical probabilistic models, providing evaluation of UAF group efficiency operation, considering UAF reliability as well as on board navigation and weapon facility performances; algorithm of UAF routes at the stages of group motion to meeting point and further movement to target area. Developed algorithm provides definition both direction and velocity of UAF, considering actual environment condition and dangerous of UAF recognition by radar, acoustic or infrared facilities of enemy; algorithms of UAF group operation scheduling at the mission target area.
Article
Unmanned aerial vehicles (UAVs) have significant prospects in a plethora of public and civic spheres. Recently, UAVs have focused primarily on applications where human presence is either impossible or hazardous. A swarm of small UAVs can cooperatively complete operations more proficiently and economically than a single large UAV. However, many issues must be resolved before stable and reliable multi-UAV networks can be realized. Task assignment in fleets of UAVs is concerned with cooperative decision-making and control. UAVs possess various functional abilities and kinematic constraints while carrying limited resources onboard. UAVs are nominated to execute multiple sequential tasks supportively on numerous ground targets. The prime objective of task assignment is to minimalize the task accomplishment time and UAV energy consumption. To date, several task assignment algorithms have been designed for UAV networks, and they are comprehensively surveyed in this paper in terms of their main ideas, operational features, advantages, and limitations. These task assignment algorithms are then compared in terms of their significant characteristics and performance factors. To the best of the authors' knowledge, no survey on task assignment techniques for different UAV missions currently exists in the literature. We also discuss open issues and challenges and then suggest projections for task assignment algorithms concerning possible future directions.
Article
Full-text available
In this paper, based on task sequence and time constraint in the SEAD mission of multi-UAV, a heterogeneous multi-UAV cooperative task assignment mathematical model is established. We put forward a hybrid algorithm GSA-GA(gravity search algorithm-genetic algorithm) to resolve cooperative task assignment. The algorithm combines gravity search algorithm and genetic algorithm, improves the coding and decoding methods in updating the position. The simulation result shows that the GSA-GA has rapid convergence rate in solving the cooperative task assignment compared with the classic DPSO algorithm, and has the better resolution. © 2018, Editorial Board of Journal of Northwestern Polytechnical University. All right reserved.
Article
Full-text available
With the development of unmanned aerial vehicle technology, unmanned aerial vehicles have been widely used in inspection of overhead power lines. Compared to traditional helicopters, unmanned aerial vehicles represent advantages of low cost and flexibility that unmanned aerial vehicles can cross complex terrain and security, more important pilots generally take no life risk during tasks. However, unmanned aerial vehicles are significantly limited by inefficiency, flying range and departure spot. Thus energy companies are seeking for a scheme that makes unmanned aerial vehicles and helicopter coordinate in inspection of overhead power lines. This paper presents a mathematic model to assess the performances of unmanned aerial vehicles and helicopter in inspection of overhead power lines in order to provide reference in seeking aerial vehicles coordination scheme.
Conference Paper
This paper proposes an algorithm for the decentralized coalition formation of multiple heterogeneous UAVs that cooperatively perform a search and attack mission to neutralize the static or dynamic ground targets in a highly uncertain region where no prior information is available about the targets, and no centralized communication link with the UAVs is possible. In such cooperative missions, if the detecting UAV does not have enough resources to neutralize a target then a coalition of UAVs may needs to be formed that fulfills the target resource requirement. This coalition formation is computationally complex due to the combinatorial nature of the problem and is NP-Hard. Therefore, solutions with low computational complexity are required. The proposed decentralized coalition formation algorithm is sub-optimal and is computationally less complex. The proposed algorithm is based on the resource match cost criteria, in which the algorithm iteratively selects the final coalition members from the responding UAVs which are closest in resource match to the target resource requirement. After the selection of every single coalition member, the new resource match cost on the basis of new target resource requirement is calculated for the remaining responding UAVs and the above procedure is repeated until the target resource requirement is satisfied. The main objective of the algorithm is that, the coalition formed must be of minimum size and closest in match to the target resource requirement, so that more UAVs and resources must remain available for the search of other targets. The proposed solution enables the UAVs to form parallel coalitions to neutralize multiple targets and to utilize their resources more effectively. The performance of the proposed algorithm is evaluated through simulation tests and the results are compared with one of the reference sub-optimal decentralized coalition formation algorithm. The results show that the proposed algorithm is more effective and determines the near optimal set of member UAVs for the decentralized coalition formation. However, the proposed algorithm is naive with less computational complexity and can produce sub-optimal solutions which in most cases are near optimal.
Chapter
A team of unmanned aerial vehicles (UAVs) with limited communication sensing ranges, communication delays, and finite resources are deployed search and destroy targets in a given region. The targets can be static or moving.
Conference Paper
This paper presents the implementation of a complete 3-D Dubins-based path planner, including non-coplanar and coplanar poses. Mathematical representation of a Dubins path in 3-dimensions has been extended into a complete planning methodology that provides a suitable solution to the implementation issues typically encountered for such a process. These include the numerical solution of the nonlinear implicit vector equations, as well as handling of non-coplanar and coplanar pose combinations. Since the suggested methodology breaks down for the case of coplanar poses, a complete planner is provided by diagnosing coplanar sets of poses, determining the solution within the plane containing them, and converting the 2-D solution to the 3-D environment.
Conference Paper
This paper presents a complete methodology associated with the implementation of the 2-dimensional clothoid path planner and trajectory generation algorithm for unmanned aerial vehicles (UAVs) applications. Based on previously suggested approaches for integrating the clothoid into the autonomous navigation and control architecture to produce a continuous curvature profile, the methodology presented in this paper has been extended to identify and provide a solution for the various issues encountered during a complete implementation of the clothoid planner for UAV trajectory generation. This includes a quadrant-based scheme for selecting the shortest path combination based upon the relative position and angle of the poses and a numerical solution of the nonlinear vector equations which define the ultimate path profile. The result is a ready-to-use planner that could easily be implemented on-board UAV hardware. The approach is demonstrated using the West Virginia University UAV Simulation Environment.
Article
Full-text available
This paper provides an algorithmic description of a solution to the task allocation problem where 'x' multiple agents, each of a given initial state are asked to transition to 'n' number of target states where n can be greater than, equal to, or less than x. The paper explains and demonstrates the methods by focusing on the specific problem of having 'x' multiple heterogeneous robotic agents travel to a of given set of 'n' target coordinates while minimizing either time or fuel in a real-time RoboFlag testbed scenario(5). The algorithm is presented as 3 modular methods based on the Primitives algorithm (6) in the first level, a Shell and K-Opt method (7) and a Genetic Programming method (8) in the second level and a new algorithm is presented in the final third level. The final level's method is compared against solving the same problem using MILP techniques as a proof of performance.
Conference Paper
This paper presents a fast algorithm for allocation at mission-time of moving targets to a group of unmanned vehicles. A fleet of UAVs must fly through a known environment to reach partially unknown locations, or targets, where three tasks: identification, attack and verification must be performed sequentially. The total mission cost is identified to be the sum of the total times that the UAVs spend completing their tasks, while respecting the task priorities and ensuring the task precedence laws. The problem is solved in two steps; the first step is performed off-line and is the most computationally intensive: the environment is subdivided into triangle-shaped areas forming the tessellation graph (TG), and the shortest path between each two vertexes couples of the plane is computed using the all-pairs-nodes Dijkstra algorithm. The second step, at mission-time, regards management of moving targets and adaptation to the results of the identification phase. Optimal task assignment is performed using the Hungarian algorithm; exact path lengths between vehicles and targets are computed from the off-line computed Dijkstra paths. One parameter is available to tune the optimal task allocation algorithm with respect to desired aggressive/selfish or cooperative behavior.
Article
Full-text available
In this paper we propose auction algorithms for solving several types of assignment problems with inequality constraints. Included are asymmetric problems with dierent numbers of persons and objects, and multiassignment problems, where persons may be assigned to several objects and reversely. A central new idea in all these algorithms is to combine regular auction, where persons bid for objects by raising their prices, with reverse auction, where objects compete for persons by essentially oering discounts. Reverse auction can also be used to accelerate substantially (and
Article
Full-text available
This paper addresses the problems of autonomous task allocation and trajectory planning for a fleet of UAVs. Two methods are compared for solving the optimization that combines task assignment, sub-jected to UAV capability constraints, and path plan-ning, subjected to dynamics, avoidance and timing constraints. Both sub-problems are non-convex and the two are strongly-coupled. The first method ex-presses the entire problem as a single mixed-integer linear program (MILP) that can be solved using available software. This method is guaranteed to find the globally-optimal solution to the problem, but is computationally intensive. The second method employs an approximation for rapid computation of the cost of many different trajectories. This enables the assignment and trajectory problems to be decou-pled and partially distributed, offering much faster computation. The paper presents several examples to compare the performance and computational re-sults from these two algorithms.
Conference Paper
Full-text available
We present a new polynomial-space algorithm, called Adopt, for distributed constraint optimization (DCOP). DCOP is able to model a large class of collaboration problems in multi agent systems where a solution within given quality parameters must be found. Existing methods for DCOP are not able to provide theoretical guarantees on global solution quality while operating both efficiently and asynchronously. Adopt is guaranteed to find an optimal solution, or a solution within a user-specified distance from the optimal, while allowing agents to execute asynchronously and in parallel. Adopt obtains these properties via a distributed search algorithm with several novel characteristics including the ability for each agent to make local decisions based on currently available information and without necessarily having global certainty. Theoretical analysis shows that Adopt provides provable quality guarantees, while experimental results show that Adopt is significantly more efficient than synchronous methods. The speedups are shown to be partly due to the novel search strategy employed and partly due to the asynchrony of the algorithm.
Article
Full-text available
In this paper we propose auction algorithms for solving several types of assignment problems with inequality constraints. Included are asymmetric problems with di#erent numbers of persons and objects, and multiassignment problems, where persons may be assigned to several objects and reversely. A central new idea in all these algorithms is to combine regular auction, where persons bid for objects by raising their prices, with reverse auction, where objects compete for persons by essentially o#ering discounts. Reverse auction can also be used to accelerate substantially (and sometimes dramatically) the convergence of regular auction for symmetric assignment problems. 1 This work was supported in part by the BM/C3 Technology branch of the United States Army Strategic Defense Command. 2 Department of Electrical Engineering and Computer Science, M. I. T., Cambridge, Mass., 02139. 3 Department of Electrical Engineering, Boston University, and ALPHATECH, Inc., Burlington, Mass., 01803. 4 ALPHATECH, Inc., Burlington, Mass., 01803. 1 1.
Article
This paper addresses the development of a hierarchical distributed control system for unmanned aerospace vehicles, specifically wide area search munitions. The hierarchy has three levels of decomposition. The top level performs task assignment using a market based bidding procedure. The middle subteam level coordinates cooperative tasks. The lower level executes the tasks, and performs trajectory optimization. The tasks, or functions, include cooperative search, cooperative classification, cooperative attack, and cooperative battle damage assessment. The control system is decentralized with no explicit leader, is redundant, and thus fault tolerant, and has on-line optimal decision capability that allows operation in a dynamic unstructured environment. Simulations are performed for a team of up to eight vehicles that show superior performance over that achievable when the vehicles operate independently.
Article
This paper surveys a new and comprehensive class of algorithms for solving the classical linear network flow problem and its various special cases such as shortest path, max-flow, assignment, transportation, and transhipment problems. The prototype method, from which the other algorithms can be derived, is the auction algorithm for the assignment problem. This is an intuitive method that operates like a rel auction where persons compete for objects by raising their prices through competitive bidding; the prices can be viewed as dual variables. Conceptually, auction algorithms represent a significant departure from the cost improvement idea that underlies primal simplex and dual ascent methods; at any one iteration, they may deteriorate both the primal and the dual cost. Auction algorithms perform very well for several important types of problems, both in theory and in practice, and they are also well suited for parallel computation.
Conference Paper
A weapon system consisting of a swarm of air vehicles whose mission is to search for, classify, attack, and perform battle damage assessment, is considered. It is assumed that the target field information is communicated to all the elements of the swarm as it becomes available. A network flow optimization problem is posed whose readily obtained solution yields the optimum resource allocation among the air vehicles in the swarm. Hence, the periodic reapplication of the centralized optimization algorithm yields the benefit of cooperative feedback control
CooperativeControlforUAVs
  • P Chandler
  • M Pachter
  • S Rasmussen
Task Allocation for Wide Area Search Munitions via Network Flow Optimization
  • C Schumacher
  • P Chandler
  • Aiaa Gnc
Optimal Task Timing using Mixed Integer Linear Programming
  • Lior S Pachter
  • M Pachter
  • P Chandler
  • C Schumacher
  • S Rasmussen