جستجوی مقالات مرتبط با کلیدواژه "moving targets" در نشریات گروه "مکانیک"

According to recent achievements in multi-agent systems, the use of cooperative arrangement of the multi-agent systems like robots and UAVs such as rescuing, aerial mapping, aerial photography, and goods delivery missions is growing as fast as possible. Implementation a cooperative mission involves two Characteristic: Assigning tasks to each agent in optimally way and path planning of agents to their related tasks. In this article, a hierarchical approach based on linear programming algorithm is developed in order to solve task assignment problem with moving targets and multiple tasks on each target, which has a non-linear structure. Dynamic targets, which have variable position and their position change in every stage of algorithm as a time varying parameter, are considered. All the timing and non-timing constraint with the consideration of flight endurance time, priorities and the moving states of each agent and target will be satisfied. The most part of forming this hierarchical approach is assuming variable position for each target and solving the dynamic task assignment by linearization and discretization of time variables and using time windows for solving stage by stage.

Recent developments in control and the use of automatic agents result in attracting attention to the cooperative systems and their abilities in performing efficient missions. Optimal task assignment of multiple cooperative agents is among the most important factors in effective performing of robots and UAVs' cooperative missions. Uncertainty and dynamic conditions are some of the most essential criteria during the simulation of mechanical phenomena.In this paper, the task assignment of a fleet of cooperative heterogeneous UAVs in a dynamic environment including moving targets was examined. An appropriate architecture was applied by using a hierarchical mixed-integer linear programming in order to solve task assignment problems within the reasonable range. To enhance the outlook of solution architecture and therefore achieving some near optimum assignments, a combination of the proposed architecture and a TSK-type Fuzzy inference system was introduced. The results approved a conflict-free and optimal solution in the task assigning problems. Efficiency increase of the proposed innovative architecture in existence of Fuzzy inference systems ranges from 10 to 35 percent. In order to evaluate the proposed approach in experimental and practical terms, a user panel has been designed and used and with the help of this panel, the capability of practical implementation of this approach has been presented.

Recent developments in mission implementation and agent control have drawn attention to the performing of cooperative missions. Decision-making systems are among the most important parts of such missions. The aim of these systems is efficient task assignment and path planning. This paper combines a task assignment approach and a path planning method to tackle collaborative problems. The task assignment approach, used in this paper, is based on an improved linear programming approach for its high accuracy. The path planning method is based on Dubins curve to handle the limitation movements of fixed-wing fighters, investigated in this paper. The proposed methods perform based on the utilized fighter’s performance characteristics and flight constraints in a dynamic environment with moving targets. The path planning approach is based on specific constraints, including certain output vectors of each point as well as input ones to the target points, minimum rotational angle, and performance factors of fighters. In the next step, Dubins path planning output is used as the necessary coefficient to solve the task assignment problem of cooperative fighters. By using a heuristic hierarchical approach based on linear programming, a comprehensive path planning and task assignment of a fleet of fighters is completed. The presented results show optimal performance and higher speed to solve rather than the classical approaches, which is capable to cover the dynamic environment as well.

In this paper، guidance laws in the direction and perpendicular to the line-of-sight is designed. The tangential component of missile acceleration is designed for zeroing the LOS rate and the other component is for controlling the closing velocity. In this method، because of using the sliding mode control، the nonlinear equation of motion is used and the acceleration command input is determined considering the target acceleration as an uncertainty. Therefore the guidance laws are robust with respect to target maneuvering and precise measuring or estimating of target acceleration is not required. In these guidance laws there are parameters for adjusting the reaching time to sliding surfaces and for producing smooth command acceleration and removing the chattering phenomenon in control input، continuous approximation method has been applied. Because of using a new sliding condition، the boundary layer is adjust only once and is independent with the designing parameter.