Evolution of fuzzy behaviors for multi-robotic system

In a multi-robotic system, robots interact with each other in a dynamically changing environment. The robots need to be intelligent both at the individual and group levels. In this paper, the evolution of a fuzzy behavior-based architecture is discussed. The behavior-based architecture decomposes the complicated interactions of multiple robots into modular behaviors at different complexity levels. The fuzzy logic approach brings in human-like reasoning to the behavior construction, selection and coordination. Various behaviors in the fuzzy behavior-based architecture are evolved by genetic algorithm (GA). At the lowest level of the architecture hierarchy, the evolved fuzzy controllers enhanced the smoothness and accuracy of the primitive robot actions. At a higher level, the individual robot behaviors have become more skillful after the evolution. At the topmost level, the evolved group behaviors have resulted in aggressive competition strategy. The simulation and real-world experimentation on a robot-soccer system justify the effectiveness of the approach.     

A flexible layered control policy for resource allocation in a sensor grid

The paper proposes a flexible layered control policy for sensor resource allocation in a sensor grid. In order to allocate sensor resources in the system to maximize the sensor grid utility, different controllers are deployed at three levels: a job-level controller, an application group controller, and a sensor grid system controller. At the lowest levels, job-level controllers perform fast, frequent, local adaptation for optimizing a single sensor grid application at a time, while, at the highest levels, sensor grid system controllers perform less frequent control actions to optimize all applications. Sensor grid system control considers all sensor grid applications in response to large system changes at coarse time granularity. Sensor grid system control exploits the interlayer coupling of the resource layer and the application layer to achieve a system-wide optimization based on the sensor grid users’ preferences. Job-level control adapts a single application to small changes at fine granularity. The layered control system uses a set of utility functions to evaluate the performance of sensor grid applications and groups. The control system chooses control actions that would result in a higher level of utility. In the simulation, a performance evaluation of the algorithm is carried out.     

A comparative study of control techniques for an underwater flight vehicle

Unmanned, underwater vehicles have been developed considerably in recent years. Remotely operated vehicles (ROVs) are increasingly used for routine inspection and maintenance tasks but have a range that is limited by the umbilical cable. For long range operations, such as oceanographic exploration and surveying, autonomous underwater vehicles (AUVs) are emerging which haveon-board power and are equipped with advanced control capabilities to carry out tasks with the minimum of human intervention. AUVs typically resemble torpedoes in that mosthave control surfaces and a single propulsion unit, and must move forwards to manoeuvre. Such vehicles are called flight vehicles. This paper describes techniques which are candidates for control of a flight AUV and identifies controllers used on some existing vehicles. Since underwater vehicle dynamics are nonlinear, fuzzy logic and sliding mode control were felt to have promise for autopilot application due to their potential robustness. Following development using a comprehensive simulation programme, the controllers were tested using the experimental vehicle, Subzero II, and their performance compared with that of a classical linear controller. The relative merits of the methods for practical implementation are discussed.     

Autonomous underwater vehicles: Hybrid control of mission and motion

This paper provides an experimental implementation and verification of a hybrid (mixed discrete state/ continuous state) controller for semi-autonomous and autonomous underwater vehicles in which the missions imply multiple task robot behavior. An overview of some of the missions being considered for this rapidly developing technology is mentioned including environmental monitoring, underwater inspection, geological survey as well as military missions in mine countermeasures.The functionalities required of such vehicles and their relation to intelligent control technology is discussed. In particular, the use of Prolog as a computer language for the specification of the discrete event system (DES) aspects of the mission control is proposed. The connections between a Prolog specification and the more common Petri Net graphical representation of a DES are made. Links are made between activation commands, transitioning signals, and the continuous state dynamic control system (DCS) responsible for vehicle stabilization.Details are given of the NPS Phoenix vehicle implementation at the present time, together with experimental validation of the concepts outlined using a simplified example mission. The paper ends with a listing of questions and concerns for the evaluation of software controllers. A list of references is given for readers interested in this subject.     

Fuzzy behavior-based control of mobile robots

An extensive fuzzy behavior-based architecture is proposed for the control of mobile robots in a multiagent environment. The behavior-based architecture decomposes the complex multirobotic system into smaller modules of roles, behaviors and actions. Fuzzy logic is used to implement individual behaviors, to coordinate the various behaviors, to select roles for each robot and, for robot perception, decision-making, and speed control. The architecture is implemented on a team of three soccer robots performing different roles interchangeably. The robot behaviors and roles are designed to be complementary to each other, so that a coherent team of robots exhibiting good collective behavior is obtained.     

Development of a real-time control architecture for a semi-autonomous underwater vehicle for intervention missions

The need for autonomous underwater vehicles (AUVs) for intervention missions becomes greater as they can perform underwater tasks requiring physical contacts with the underwater environment, such as underwater plug-in/plug-out, construction and repair, cable streaming, mine hunting, munitions retrieval, and scientific sampling. This paper describes a semi-autonomous underwater vehicle for intervention missions that has multiple on-board CPUs, redundant sensors and actuators, on-board power source and a robotic manipulator for dextrous underwater performance. Such a complex robotic vehicle system requires advanced control software architecture for on-board intelligence with a wide range of sensors and actuators to carry out required missions. In this paper, AUV control architectures are reviewed and a sensor data bus based control architecture (SDBCA) is presented. SDBCA is a modified hierarchical architecture that offers good controllability and stability while sensor data bus increases flexibility of system design, making it possible to have a prompt response from high-level control with respect to low-level sensor data. The overall sensor input mechanism of SDBCA becomes similar to the sensor input mechanism of subsumption architecture.     

Distributed fuzzy discrete event system for robotic sensory information processing

Abstract: This paper presents a novel intelligent sensory information processing technique using a fuzzy discrete event system (FDES) for robotic control. The proposed method combines the predictive control approach of a discrete event system with the approximate reasoning aspect of fuzzy logic. It develops a supervisory control strategy for behavior-based robotic control using distributed FDES. The application of distributed FDES eliminates the formation of complex fuzzy predicates and a large fuzzy rule-base. The FDES-based approach also provides means for analyzing behavior-based decision-making using the observability and controllability of an FDES. The observability of an FDES describes uncertainties in sensory data, and the controllability of an FDES exploits uncertain state transitions in a dynamic environment. Comprehensive experiments on behavior-based mobile robot navigation are presented to authenticate the performance of the proposed methodology.     

Adaptive Fuzzy Sliding Mode Controller for the Kinematic Variables of an Underwater Vehicle

This paper address the kinematic variables control problem for the low-speed manoeuvring of a low cost and underactuated underwater vehicle. Control of underwater vehicles is not simple, mainly due to the non-linear and coupled character of system equations, the lack of a precise model of vehicle dynamics and parameters, as well as the appearance of internal and external perturbations. The proposed methodology is an approach included in the control areas of non-linear feedback linearization, model-based and uncertainties consideration, making use of a pioneering algorithm in underwater vehicles. It is based on the fusion of a sliding mode controller and an adaptive fuzzy system, including the advantages of both systems. The main advantage of this methodology is that it relaxes the required knowledge of vehicle model, reducing the cost of its design. The described controller is part of a modular and simple 2D guidance and control architecture. The controller makes use of a semi-decoupled non-linear plant model of the Snorkel vehicle and it is compounded by three independent controllers, each one for the three controllable DOFs of the vehicle. The experimental results demonstrate the good performance of the proposed controller, within the constraints of the sensorial system and the uncertainty of vehicle theoretical models.     

Fuzzy-Logic Based Navigation of Underwater Vehicles

A fuzzy logic based general purpose modular control architecture is presented for underwater vehicle autonomous navigation, control and collision avoidance. Three levels of fuzzy controllers comprising the sensor fusion module, the collision avoidance module and the motion control module are derived and implemented. No assumption is made on the specific underwater vehicle type, on the amount of a priori knowledge of the 3-D undersea environment or on static and dynamic obstacle size and velocity. The derived controllers account for vehicle position accuracy and vertical stability in the presence of ocean currents and constraints imposed by the roll motion. The main advantage of the proposed navigation control architecture is its simplicity, modularity, expandability and applicability to any type of autonomous or semi-autonomous underwater vehicles. Extensive simulation studies are performed on the NPS Phoenix vehicle whose dynamics have been modified to account for roll stability.     

Hierarchical control of aircraft propulsion systems: Discrete event supervisor approach

This paper presents an application of the recently developed theory of language-measure-based discrete event supervisory (DES) control to aircraft propulsion systems. A two-layer hierarchical architecture is proposed to coordinate the operations of a twin-engine propulsion system. The two engines are individually controlled to achieve enhanced performance and reliability, as necessary for fulfilling the mission objectives. Each engine, together with its continuously varying control system, is operated at the lower level under the supervision of a local discrete-event controller for condition monitoring and life extension; the gain-scheduled feedback controller that is used to maintain the specified performance of the turbofan engine is kept unaltered. A global DES controller at the upper level coordinates the local DES controllers for load balancing and health management of the propulsion system.     

A comparative investigation of non-linear activation functions in neural controllers for search-based game AI engineering

The creation of intelligent video game controllers has recently become one of the greatest challenges in game artificial intelligence research, and it is arguably one of the fastest-growing areas in game design and development. The learning process, a very important feature of intelligent methods, is the result of an intelligent game controller to determine and control the game objects behaviors’ or actions autonomously. Our approach is to use a more efficient learning model in the form of artificial neural networks for training the controllers. We propose a Hill-Climbing Neural Network (HillClimbNet) that controls the movement of the Ms. Pac-man agent to travel around the maze, gobble all of the pills and escape from the ghosts in the maze. HillClimbNet combines the hill-climbing strategy with a simple, feed-forward artificial neural network architecture. The aim of this study is to analyze the performance of various activation functions for the purpose of generating neural-based controllers to play a video game. Each non-linear activation function is applied identically for all the nodes in the network, namely log-sigmoid, logarithmic, hyperbolic tangent-sigmoid and Gaussian. In general, the results shows an optimum configuration is achieved by using log-sigmoid, while Gaussian is the worst activation function.     

DEVS-BASED INTELLIGENT CONTROL: SPACE-ADAPTED MIXING SYSTEM EXAMPLE

This paper reviews a methodology for event-based intelligent control employing the DEVS (discrete event system specification) formalism. In this control paradigm, the controller expects to receive confirming sensor responses to its control commands within definite time windows determined by its DEVS model of the system under control. We apply the DEVS-based intelligent control paradigm to a space-adapted mixing system. The event-based approach is compared with conventional sequential control methods.     

Balanced reactive-deliberative architecture for multi-agent system for simulation league of RoboCup

This paper presents an architecture for a multi-agent system for the RoboCup simulation league. It consists of a dynamic dual behavior-based architecture for an intelligent agent, a behavior-based decision algorithm, and a dynamic role-based multi-agent cooperation model. A new concept called confidence function is introduced to balance reactivity and deliberation. This architecture was implemented in a team, and match results demonstrate its validity.     

自治水下机器人的研发现状与展望

本文介绍了自治水下机器人的研究意义和分类定义.对水下机器人的研究现状和国内外有代表性的自治水下机器人型号进行了综述.讨论分析了水下机器人研究的的关键技术,包括智能水平,能源问题,导航系统,传感器系统,水下通讯技术和多水下机器人协作研究,并指出了水下机器人的未来发展方向.     

Asymptotic tracking by a reinforcement learning-based adaptive critic controller

Adaptive critic (AC) based controllers are typically discrete and/or yield a uniformly ultimately bounded stability result because of the presence of disturbances and unknown approximation errors. A continuous-time AC controller is developed that yields asymptotic tracking of a class of uncertain nonlinear systems with bounded disturbances. The proposed AC-based controller consists of two neural networks (NNs) – an action NN, also called the actor, which approximates the plant dynamics and generates appropriate control actions; and a critic NN, which evaluates the performance of the actor based on some performance index. The reinforcement signal from the critic is used to develop a composite weight tuning law for the action NN based on Lyapunov stability analysis. A recently developed robust feedback technique, robust integral of the sign of the error (RISE), is used in conjunction with the feedforward action neural network to yield a semiglobal asymptotic result. Experimental results are provided that illustrate the performance of the developed controller.     

Control synthesis of timed discrete event systems based onpredicate invariance

In this paper, arc-timed Petri nets are used to model controlled real-time discrete event systems, and the control synthesis problem that designs a controller for a system to satisfy its given closed-loop behavior specification is addressed. For the problem with the closed-loop behavior specified by a state predicate, real-time control-invariant predicates are introduced, and a fixpoint algorithm to compute the unique extremal control-invariant subpredicate of a given predicate, key to the control synthesis, is presented. For the problem with the behavior specified by a labeled arc-timed Petri net, it is shown that the control synthesis problem can be transformed into one that synthesizes a controller for an induced arc-timed Petri net with a state predicate specification. The problem can then be solved by using the fixpoint algorithm as well. The algorithm involves conjunction and disjunction operations of polyhedral sets and can be algorithmically implemented, making automatic synthesis of controllers for real-time discrete event systems possible.     

Application of networked discrete event system theory on intelligent transportation systems

The responses of vehicles to the changes in traffic situations inevitably have delays in observing an event and implementing a control command, which often causes fatal accidents. So far, the methods for handling delays are empirical and cannot be mathematically proven. To eliminate the accidents caused by such delays, in this paper, we develop mathematically provable methods to handle these delays. Specifically, we use networked discrete event systems to model the process of driving vehicles and present a supervisory controller for handling delay situations. The method developed in this paper could serve as a new start for modeling and controlling the responsive behaviors of self-driving vehicles in the future.     

一种基于行为控制的两自由度机械臂智能控制器

基于行为的控制方法相对于传统的控制方法在解决未知环境中的机器人中有着更好的鲁棒性和实时性.本文提出了一种基于反应式行为控制的智能控制器,以强化学习作为智能控制器的学习算法.通过采用评价-控制模型,该智能控制器能够不依赖于系统模型,通过连续地在线学习得到机器人的行为.将该智能控制器应用到两自由度仿真机械臂的控制中,仿真结果表明该智能控制器可以实现对两自由度机械臂的连续控制,使其能够迅速达到目标位置.     

On controllability of discrete event systems in a behavioral framework

The purpose of this paper is to investigate the controllability and the achievability of discrete event systems within a behavioral framework. Based on the notion of Willems’ behavioral controllability [1, 2], we introduce a new concept related to the controllability of discrete event systems. By using the controllability proposed here and the notion related to achievable behaviors [3, 4], we show that the behavioral controllability for a given specification with respect to language is equivalent to the existence of a controller, so that an interconnected system satisfies the specification exactly. A proposed controller here is represented by the intersection of the behavior of a given plant and that of a given (controllable) specification. We also clarify that our controllers for a given specification fit the properties of well-known supervisory controllers proposed and developed by Ramadge and Wonham [5]. The text was submitted by the authors in English.     

Energy optimal depth control for long range underwater vehicles with applications to a hybrid underwater glider

The efficient control of a vehicle’s depth is a ubiquitous need of underwater vehicles for long term deployments. Using standard hydrodynamic drag and lift formulations for a hybrid autonomous underwater glider (AUG) the performance penalty for long range underwater vehicles with a non zero buoyant force is shown to be significant. To this end, an energy optimal depth controller design methodology for a long range autonomous underwater vehicle is presented with applications to a propeller driven hybrid AUG during level flight. The method makes use of a reduced order linear model that has been validated from field data. The standard state space model is augmented with the state integral matrix and rewritten to the state error representation. The resulting representation is well suited to the computation of energy optimal gains for a linear quadratic regulator. Field demonstrations of the standard pitching depth controller and the energy optimal depth controller using the computed gains shows the reduced order model to be sufficient for the purpose of the controller design, improving on the standard pitching depth controller response, transport efficiency and ease of tuning.