基于视觉的移动机器人可通行区域识别研究综述

实现实时准确的可通行区域识别,是户外环境下移动机器人导航的重要组成部分。对基于视觉的移动机器人可通行区域识别研究进行了综述,首先介绍了移动机器人视觉导航常用的视觉系统,并从障碍物检测、地形分类两个方面介绍了该问题研究的进展,最后对该领域的技术发展趋势进行了探讨。     

A biologically inspired method for vision-based docking of wheeled mobile robots

We present a new control law for the problem of docking a wheeled robot to a target at a certain location with a desired heading. Recent research into insect navigation has inspired a solution which uses only one video camera. The control law is of the “behavioral” type in that all control actions are based on immediate visual information. Docking success under certain conditions is proved mathematically and simulation studies show the control law to be robust to camera intrinsic parameter errors. Experiments were performed for verification of the control law.     

Data association and occlusion handling for vision-based people tracking by mobile robots

This paper presents an approach for tracking multiple persons on a mobile robot with a combination of colour and thermal vision sensors, using several new techniques. First, an adaptive colour model is incorporated into the measurement model of the tracker. Second, a new approach for detecting occlusions is introduced, using a machine learning classifier for pairwise comparison of persons (classifying which one is in front of the other). Third, explicit occlusion handling is incorporated into the tracker. The paper presents a comprehensive, quantitative evaluation of the whole system and its different components using several real world data sets.     

移动机器人Java Agent控制系统设计

针对移动机器人的任务和硬件组成,提出了基于Java 开发平台的Agent控制系统设计方法。以目前应用较广泛的JADE作为Agent开发平台,采用JNI方法实现了Agent与硬件系统的交互。在运动控制卡上设计了有实时性要求的轨迹生成、运动控制、位姿估计和安全控制等4个行为任务,将数据库和路径规划等管理性行为设计在车载PC104工业控制计算机上。人机交互界面可作为独立的Agent驻留在上位监控计算机上。这种方法结合了Java Agent开发平台的普遍性和工业控制的实时性,实验证明了该方法的可行性。     

Cooperative behavior acquisition for mobile robots in dynamically changing real worlds via vision-based reinforcement learning and development

In this paper, we first discuss the meaning of physical embodiment and the complexity of the environment in the context of multi-agent learning. We then propose a vision-based reinforcement learning method that acquires cooperative behaviors in a dynamic environment. We use the robot soccer game initiated by RoboCup (Kitano et al., 1997) to illustrate the effectiveness of our method. Each agent works with other team members to achieve a common goal against opponents. Our method estimates the relationships between a learner's behaviors and those of other agents in the environment through interactions (observations and actions) using a technique from system identification. In order to identify the model of each agent, Akaike's Information Criterion is applied to the results of Canonical Variate Analysis to clarify the relationship between the observed data in terms of actions and future observations. Next, reinforcement learning based on the estimated state vectors is performed to obtain the optimal behavior policy. The proposed method is applied to a soccer playing situation. The method successfully models a rolling ball and other moving agents and acquires the learner's behaviors. Computer simulations and real experiments are shown and a discussion is given.     

Learning control of mobile robots using a multiprocessor system

A real-time multiprocessor system is proposed for the solution of the tracking problem of mobile robots operating in a real context with environmental disturbances and parameter uncertainties. The proposed control scheme utilizes multiple models of the robot for its identification in an adaptive and learning control framework. Radial Basis Function Networks (RBFNs) are considered for the multiple models in order to exploit the net non-linear approximation capabilities for modeling the kinematic behavior of the vehicle and for reducing unmodeled contributions to tracking errors. The training of the nets and the tests of the achieved control performance have been done in a real experimental setup. The proposed control architecture improves the robot tracking performance achieving fast and accurate control actions in presence of large and time-varying uncertainties in dynamical environments. The experimental results are satisfactory in terms of tracking errors and computational efforts.     

一类有序化多移动机器人群集运动控制系统

群集运动控制(flocking control)是一种新型的多移动机器人运动协调控制, 目前的研究多集中于无leader模式下群集运动控制器的设计. 为此, 本文阐述了一类多移动机器人有序化群集运动系统控制方案及其性能评价方法. 首先, 在前人的研究基础上, 本文介绍了基于Agent的有序化编队控制机制; 然后, 运用非完整约束下移动机器人的动力学原理, 设计了由Agent到移动机器人的控制转化方法; 并进一步提出了基于“最小稳定时间”的群集运动分析法, 可对有序化群集运动系统进行分析; 最后, 运用仿真实例, 描述了多移动机器人有序化群集运动的控制及分析过程. 实验结果验证了此控制方案的有效性.     

Integrated tracking and accident avoidance system for mobile robots

In the intelligent transportation field, various accident avoidance techniques have been applied. One of the most common issues with these is the collision, which remains an unsolved problem. To this end, we developed a Collision Warning and Avoidance System (CWAS), which was implemented in the wheeled mobile robot. Path planning is crucial for a mobile robot to perform a given task correctly. Here, a tracking system for mobile robots that follow an object is presented. Thus, we implemented an integrated tracking system and CWAS in a mobile robot. Both systems can be activated independently. Using the CWAS, the robot is controlled through a remotely controlled device, and collision warning and avoidance functions are performed. Using the tracking system, the robot performs tasks autonomously and maintains a constant distance from the followed object. Information on the surroundings is obtained through range sensors, and the control functions are performed through the microcontroller. The front, left, and right sensors are activated to track the object, and all the sensors are used for the CWAS. The proposed system was tested using the binary logic controller and the Fuzzy Logic Controller (FLC). The efficiency of the robot was improved by increasing the smoothness of motion via the FLC, achieving accuracy in tracking and increasing the safety of the CWAS. Finally, simulations and experimental outcomes have shown the usefulness of the system.     

Detection of doors using a genetic visual fuzzy system for mobile robots

Doors are common objects in indoor environments and their detection can be used in robotic tasks such as map-building, navigation and positioning. This work presents a new approach to door-detection in indoor environments using computer vision. Doors are found in gray-level images by detecting the borders of their architraves. A variation of the Hough Transform is used in order to extract the segments in the image after applying the Canny edge detector. Features like length, direction, or distance between segments are used by a fuzzy system to analyze whether the relationship between them reveals the existence of doors. The system has been designed to detect rectangular doors typical of many indoor environments by the use of expert knowledge. Besides, a tuning mechanism based on a genetic algorithm is proposed to improve the performance of the system according to the particularities of the environment in which it is going to be employed. A large database of images containing doors of our building, seen from different angles and distances, has been created to test the performance of the system before and after the tuning process. The system has shown the ability to detect rectangular doors under heavy perspective deformations and it is fast enough to be used for real-time applications in a mobile robot.     

Development of the residual power prediction system for mobile robots

The article presents a multiple residual power prediction system to be applied in mobile robots or automation fields. The system contains multiple power detection units to measure multiple online power values. Each power detection unit uses four current sensors to measure the current variety, and uses weighted average method and redundant management method to calculate the exact current value, and isolates faulty measurement values. We use the proposed algorithms to be applied in voltage detection of each power detection unit. Then we can calculate the real-time power values according to the current and voltage measurement values. The control core of the power detection unit is HOLTEK microchip, and communicates with the data integration unit via wire I2C interface. The power detection units transmit the measurement values of current and voltage to the controller of the system. The main controller of the power detection system is PC-based system, and communicates with the data integration unit via wire RS232 interface. The main controller of the system can controls each power output ratio of the power sources, and predicts the power loading and the residual power for each power detection unit using auto-regression algorithm, and computes the residual time of mobile robots to work in the free space, and arrange the residual power of the enough power source to the weakness power sources using sequential single-item auction algorithm. In experiment result, the residual power prediction system can adjusts the power sources to increase the working time.     

Kinematic calibration for collaborative robots on a mobile platform using motion capture system

For modern robotic applications that go beyond the typical industrial environment, absolute accuracy is one of the key properties that make this possible. There are several approaches in the literature to improve robot accuracy for a typical industrial robot mounted on a fixed frame. In contrast, there is no method to improve robot accuracy when the robot is mounted on a mobile base, which is typical for collaborative robots. Therefore, in this work, we proposed and analyzed two approaches to improve the absolute accuracy of the robot mounted on a mobile platform using an optical measurement system. The first approach is based on geometric operations used to calculate the rotation axes of each joint. This approach identifies all rotational axes, which allows the calculation of the Denavit–Hartenberg (DH) parameters and thus the complete kinematic model, including the position and orientation errors of the robot end-effector and the robot base. The second approach to parameter estimation is based on optimization using a set of joint positions and end-effector poses to find the optimal DH parameters. Since the robot is mounted on a mobile base that is not fixed, an optical measurement system was used to dynamically and simultaneously measure the position of the robot base and the end-effector. The performance of the two proposed methods was analyzed and validated on a 7-DoF Franka Emika Panda robot mounted on a mobile platform PAL Tiago-base. The results show a significant improvement in absolute accuracy for both proposed approaches. By using the proposed approach with the optical measurement system, we can easily automate the estimation of robot kinematic parameters with the aim of improving absolute accuracy, especially in applications that require high positioning accuracy.     

An autonomous vision-based mobile robot

This paper describes the theoretical development and experimental implementation of a complete navigation procedure for use in an autonomous mobile robot for structured environments. Estimates of the vehicle's position and orientation are based on the rapid observation of visual cues located at discrete positions within the environment. The extended Kalman filter is used to combine these visual observations with sensed wheel rotations to produce optimal estimates continuously. The complete estimation procedure, as well as the control algorithm, developed are time independent. A naturally suitable quantity involving wheel rotations is used as the independent variable. One consequence of this choice is that the vehicle speed can be specified independently of the estimation and control algorithms. Reference paths are “taught” by manually leading the vehicle through the desired path. Estimates produced by the extended Kalman filter during this teaching session are then used to represent the geometry of the path. The tracking of taught reference paths is accomplished by controlling the position and orientation of the vehicle relative to the reference path. Time-independence path tracking has necessitated the development of a novel, geometry-based means for advancing along the reference path     

Development of a real-life EKF based SLAM system for mobile robots employing vision sensing

Developing real-life solutions for implementation of the simultaneous localization and mapping (SLAM) algorithm for mobile robots has been well regarded as a complex problem for quite some time now. Our present work demonstrates a successful real implementation of extended Kalman filter (EKF) based SLAM algorithm for indoor environments, utilizing two web-cam based stereo-vision sensing mechanism. The vision-sensing mechanism is a successful development of a real algorithm for image feature identification in frames grabbed from continuously running videos on two cameras, tracking of these identified features in subsequent frames and incorporation of these landmarks in the map created, utilizing a 3D distance calculation module. The system has been successfully test-run in laboratory environments where the robot is commanded to navigate through some specified waypoints and create a map of its surrounding environment. Our experimentations showed that the estimated positions of the landmarks identified in the map created closely tallies with the actual positions of these landmarks in real-life.     

An autonomous stereovision-based navigation system (ASNS) for mobile robots

Recently, stereovision has appeared in robotics as a source of information for real-time mapping and path planning. In this paper, an intelligent motion system for mobile robots is designed and implemented using stereovision. The proposed system uses stereovision as a primary method for sensing the environment, and the system is able to navigate intelligently in an indoor environment with varying degrees of obstacle complexity. It creates noiseless and high-confidence 3D point clouds and uses these point clouds as an input for the mapping and path-planning modules. The proposed system was built by developing, enhancing, and integrating various techniques, modules and algorithms. The Stereovision-based Path-planning module is the integration of three main enhanced techniques: (1) the multi-baseline multi-view stereovision filter (MMSVF), (2) accurate floor detection and segmentation (AFDS), and (3) the intelligent gazing module (IGM). This Stereovision-based Path planning (MMSVF, IGM, and AFDS) was integrated with the Fuzzy Logic Motion Controller (FLMC). All techniques, modules and algorithms are implemented using a multi-threaded and client–server-based architecture. To prove the viability and robustness of our proposed system, we have integrated all components of the system into a fully functional mobile robot navigation system. We compared the performance of the main modules with that of similar modules in the literatures, and showed that our modules had better performance. Testing the whole system is more important than just testing each module individually. To the best of our knowledge, the literatures lack such testing. Hence, in this paper we present the performance of our complete integrated system in different environments using different parameters and different architectures.     

Path following with a security margin for mobile robots

This paper addresses the problem of determining a feedback control law, robust with respect to localization errors, allowing a mobile robot to follow a prescribed path. The model that we consider is a dynamic extension of the usual kinematic model of a mobile robot in the sense that the path curvature is defined as a new state variable. The control variables are the linear velocity and the derivative of the curvature. By defining a sliding manifold we determine a stabilizing controller for the nominal system, that is when the exact configuration is supposed to be known. Then, we prove that the system remains stable when the feedback control inputs use estimated values instead of the exact values, and we characterize the control robustness with respect to localization and curvature estimation errors. The control robustness is expressed by determining a bounded attractive domain containing the configuration error as the closed-loop control is performed with the estimated state values. Two control laws are successively proposed. The former is deduced from Lyapunov's direct method, and the latter is based on variable structure control techniques. Using variable structure control we show that the size of the attractive domain can be easily minimized while keeping the balance between short response time, low output oscillation, and large stability domain. Knowledge of this attractive domain allows us to compute easily a security margin to guarantee obstacle avoidance during the path following process. Experimental results are presented at the end of the paper.     

Stability analysis of a vision-based control design for an autonomous mobile robot

We propose a simple control design allowing a mobile robot equipped with a camera to track a line on the ground. The control algorithm, as well as the image-processing algorithm, are very simple. We discuss the existence and the practical stability of an equilibrium trajectory of the robot when tracking a circular reference line. We then give a complementary analysis for arbitrary reference lines with bounded curvature. Experimental results confirm the theoretical analysis.     

Fault-tolerant flocking for a group of autonomous mobile robots

Consider a system composed of mobile robots that move on the plane, each of which independently executing its own instance of an algorithm. Given a desired geometric pattern, the flocking problem consists in ensuring that the robots form this pattern and maintain it while moving together on the plane. In this paper, we explore flocking in the presence of faulty robots, where the desired pattern is a regular polygon. We propose a distributed fault tolerant flocking algorithm assuming a semi-synchronous model with a k-bounded scheduler, in the sense that no robot is activated no more than k times between any two consecutive activations of any other robot.The algorithm is composed of three parts: failure detector, ranking assignment, and flocking algorithm. The role of the rank assignment is to provide a persistent and unique ranking for the robots. The failure detector identifies the set of currently correct robots in the system. Finally, the flocking algorithm handles the movement and reconfiguration of the flock, while maintaining the desired shape. The difficulty of the problem comes from the combination of the three parts, together with the necessity to prevent collisions and allow the rotation of the flock. We formally prove the correctness of our proposed solution.