Samuel B. Lazarus, Antonios Tsourdos, Brian A. White, Peter Silson, Al Savvaris, Camille‐Alain Rabbath and Nicolas Lèchevin
This paper aims to describe a recently proposed algorithm in terrain‐based cooperative UAV mapping of the unknown complex obstacle in a stationary environment where the complex…
Abstract
Purpose
This paper aims to describe a recently proposed algorithm in terrain‐based cooperative UAV mapping of the unknown complex obstacle in a stationary environment where the complex obstacles are represented as curved in nature. It also aims to use an extended Kalman filter (EKF) to estimate the fused position of the UAVs and to apply the 2‐D splinegon technique to build the map of the complex shaped obstacles. The path of the UAVs are dictated by the Dubins path planning algorithm. The focus is to achieve a guaranteed performance of sensor based mapping of the uncertain environments using multiple UAVs.
Design/methodology/approach
An extended Kalman filter is used to estimate the position of the UAVs, and the 2‐D splinegon technique is used to build the map of the complex obstacle where the path of the UAVs are dictated by the Dubins path planning algorithm.
Findings
The guaranteed performance is quantified by explicit bounds of the position estimate of the multiple UAVs for mapping of the complex obstacles using 2‐D splinegon technique. This is a newly proposed algorithm, the most efficient and a robust way in terrain based mapping of the complex obstacles. The proposed method can provide mathematically provable and performance guarantees that are achievable in practice.
Originality/value
The paper describes the main contribution in mapping the complex shaped curvilinear objects using the 2‐D splinegon technique. This is a new approach where the fused EKF estimated positions are used with the limited number of sensors' measurements in building the map of the complex obstacles.
Details
Keywords
Yueheng Qiu, Weiguo Zhang, Xiaoxiong Liu and Pengxuan Zhao
The purpose of this paper is to present the research into fault detection and isolation (FDI) and evaluation of the reduction of performance after failures occurred in the flight…
Abstract
Purpose
The purpose of this paper is to present the research into fault detection and isolation (FDI) and evaluation of the reduction of performance after failures occurred in the flight control system (FCS) during its mission operation.
Design/methodology/approach
The FDI is accomplished via using the multiple models scheme which is developed based on the Extend Kalman Filter (EKF) algorithm. Towards this objective, the healthy mode of the FCS under different type of failures, including the control surfaces and structural, should be considered. It developed a bank of extended multiple models adaptive estimation (EMMAE) to detect and isolate the above mentioned failures in the FCS. In addition, the performances including the flight envelope, the voyage and endurance in cruising are proposed to reference and evaluate the process of mission, especially for UAV under failure conditions.
Findings
The contribution of this paper is to provide the information not only about the failures, but also considering whether the UAV can accomplish the task for the ground station.
Originality/value
The main contribution of this paper is in the areas of the structural and control surface faults researching, which are occurred in the mission procedures and emphasized the identification of those failures' magnitudes. The FDI scheme includes the performance evaluation, while the evaluation obtained through the extensive numerical simulations and saved in the offline database. As a consequence, it is more accurate and less computationally demanding while evaluating the performance.
Details
Keywords
This purpose of this paper is to provide an overview of the theoretical background and applications of inverse reinforcement learning (IRL).
Abstract
Purpose
This purpose of this paper is to provide an overview of the theoretical background and applications of inverse reinforcement learning (IRL).
Design/methodology/approach
Reinforcement learning (RL) techniques provide a powerful solution for sequential decision making problems under uncertainty. RL uses an agent equipped with a reward function to find a policy through interactions with a dynamic environment. However, one major assumption of existing RL algorithms is that reward function, the most succinct representation of the designer's intention, needs to be provided beforehand. In practice, the reward function can be very hard to specify and exhaustive to tune for large and complex problems, and this inspires the development of IRL, an extension of RL, which directly tackles this problem by learning the reward function through expert demonstrations. In this paper, the original IRL algorithms and its close variants, as well as their recent advances are reviewed and compared.
Findings
This paper can serve as an introduction guide of fundamental theory and developments, as well as the applications of IRL.
Originality/value
This paper surveys the theories and applications of IRL, which is the latest development of RL and has not been done so far.
Details
Keywords
Cem Şafak Şahin and M. Ümit Uyar
This paper aims to present an approach for a bio‐inspired decentralization topology control mechanism, called force‐based genetic algorithm (FGA), where a genetic algorithm (GA…
Abstract
Purpose
This paper aims to present an approach for a bio‐inspired decentralization topology control mechanism, called force‐based genetic algorithm (FGA), where a genetic algorithm (GA) is run by each holonomic autonomous vehicle (HAV) in a mobile ad hoc network (MANET) as software agent to achieve a uniform spread of HAVs and to provide a fully connected network over an unknown geographical terrain. An HAV runs its own FGA to decide its next movement direction and speed based on local neighborhood information, such as obstacles and the number of neighbors, without a centralized control unit or global knowledge.
Design/methodology/approach
The objective function used in FGA is inspired by the equilibrium of the molecules in physics where each molecule tries to be in the balanced position to spend minimum energy to maintain its position. In this approach, a virtual force is assumed to be applied by the neighboring HAVs to a given HAV. At equilibrium, the aggregate virtual force applied to an HAV by its neighbors should sum up to zero. If the aggregate virtual force is not zero, it is used as a fitness value for the HAV. The value of this virtual force depends on the number of neighbors within the communication range of Rcom and the distance among them. Each chromosome in our GA‐based framework is composed of speed and movement direction. The FGA is independently run by each HAV as a topology control mechanism and only utilizes information from neighbors and local terrain to make movement and speed decisions to converge towards a uniform distribution of HAVs. The authors developed an analytical model, simulation software and several testbeds to study the convergence properties of the FGA.
Findings
The paper finds that coverage‐centric, bio‐inspired, mobile node deployment algorithm ensures effective sensing coverage for each mobile node after initial deployment. The FGA is also an energy‐aware self‐organization framework since it reduces energy consumption by eliminating unnecessary excessive movements. Fault‐tolerance is another important feature of the GA‐based approach since the FGA is resilient to losses and malfunctions of HAVs. Furthermore, the analytical results show that the authors' bio‐inspired approach is effective in terms of convergence speed and area coverage uniformity. As seen from the experimental results, the FGA delivers promising results for uniform autonomous mobile node distribution over an unknown geographical terrain.
Originality/value
The proposed decentralized and bio‐inspired approach for autonomous mobile nodes can be used as a real‐time topology control mechanism for commercial and military applications since it adapts to local environment rapidly but does not require global network knowledge.
Details
Keywords
Sajjad Shoja Majidabad and Heydar Toosian Shandiz
The purpose of this paper is to develop sliding mode control with linear and nonlinear manifolds in discrete‐time domain for robot manipulators.
Abstract
Purpose
The purpose of this paper is to develop sliding mode control with linear and nonlinear manifolds in discrete‐time domain for robot manipulators.
Design/methodology/approach
First, a discrete linear sliding mode controller is designed to an n‐link robot based on Gao's reaching law. In the second step, a discrete terminal sliding mode controller is developed to design a finite time and high precision controller. The stability analysis of both controllers is presented in the presence of model uncertainties and external disturbances. Finally, sampling time effects on the continuous‐time system outputs and sliding surfaces are discussed.
Findings
Computer simulations on a three‐link SCARA robot show that the proposed controllers are robust against model uncertainties and external disturbance. It was also shown that the sampling time has important effects on the closed loop system stability and convergence.
Practical implications
The proposed controllers are low cost and easily implemented in practice in comparison with continuous‐time ones.
Originality/value
The novelty associated with this paper is the development of an approach to finite time and robust control of n‐link robot manipulators in discrete‐time domain. Also, obtaining an upper bound for the sampling time is another contribution of this work.
Details
Keywords
Jacques Penders and Lyuba Alboul
This paper aims to discuss traffic patterns generated by swarms of robots while commuting to and from a base station.
Abstract
Purpose
This paper aims to discuss traffic patterns generated by swarms of robots while commuting to and from a base station.
Design/methodology/approach
The paper adopts a mathematical evaluation and robot swarm simulation. The swarm approach is bottom‐up: designing individual agents the authors are looking for emerging group behaviour patterns. Examples of group behaviour patterns are human‐driven motorized traffic which is rigidly structured in two lanes, while army ants develop a three‐lane pattern in their traffic. The authors copy army ant characteristics onto their robots and investigate whether the three lane traffic pattern may emerge. They follow a three‐step approach. The authors first investigate the mathematics and geometry of cases occurring when applying the artificial potential field method to three “perfect” robots. Any traffic pattern (two, three or more lanes) appears to be possible. Next, they use the mathematical cases to study the impact of limited visibility by defining models of sensor designs. In the final step the authors implement ant inspired sensor models and a trail following mechanism on the robots in the swarm and explore which traffic patterns do emerge in open space as well as in bounded roads.
Findings
The study finds that traffic lanes emerge in the swarm traffic; however the number of lanes is dependent on the initial situation and environmental conditions. Intrinsically the applied robot models do not determine a specific number of traffic lanes.
Originality/value
The paper presents a method for studying and simulating robot swarms.
Details
Keywords
Jingmei Zhang, Changyin Sun and Yiqing Huang
The purpose of this paper is to propose a robust control scheme for near space vehicle's (NSV's) reentry attitude tracking problem under aerodynamic parameter variations and…
Abstract
Purpose
The purpose of this paper is to propose a robust control scheme for near space vehicle's (NSV's) reentry attitude tracking problem under aerodynamic parameter variations and external disturbances.
Design/methodology/approach
The robust control scheme is composed of dynamic surface control (DSC) and least squares support vector machines (LS‐SVM). DSC is used to design a nonlinear controller for HSV; then, to increase the robustness and improve the control performance of the controller. LS‐SVM is presented to estimate the lumped uncertainties, including aerodynamic parameter variations and external disturbances. The stability analysis shows that all closed‐loop signals are bounded, with output tracking error and estimate error of LS‐SVM weights exponentially converging to small compacts.
Findings
Simulation results demonstrate that the proposed method is effective, leading to promising performance.
Originality/value
First, a robust control scheme composed of DSC and adaptive LS‐SVM is proposed for NSV's reentry attitude tracking problem under aerodynamic parameter variations and external disturbances; second, the proposed method can achieve more favorable tracking performances than conventional dynamic surface control because of employing LS‐SVM to estimate aerodynamic parameter variations and external disturbances.
Details
Keywords
Rodrigo Martins, Francisco Fernandes, Virginia Infante and Antonio R. Andrade
The purpose of this paper is to describe an integer linear programming model to schedule the maintenance crew and the maintenance tasks in a bus operating company.
Abstract
Purpose
The purpose of this paper is to describe an integer linear programming model to schedule the maintenance crew and the maintenance tasks in a bus operating company.
Design/methodology/approach
The proposed methodology relies on an integer linear programming model that finds feasible maintenance schedules. It minimizes the costs associated with maintenance crew and the costs associated with unavailability. The model is applied in a real-world case study of a Portuguese bus operating company. A constructive heuristic approach is put forward, based on solving the maintenance scheduling problem for each bus separately.
Findings
The heuristic finds better solutions than the exact methods (based on branch-and-bound techniques) in a much lower computational time.
Practical implications
The results suggest the relevance of such heuristic approaches for maintenance scheduling in practice.
Originality/value
This proposed model is an effective decision-making support method that provides feasible maintenance schedules for the maintenance technicians and for the maintenance tasks in a fleet of buses. It also complies with several operational, technical and labour constraints.