基于多层多道焊的弧焊机器人视觉信息提取及处理

介绍爬行式弧焊机器人三维视觉信息的传感原理。研究了拉普拉斯锐化、细化等措施对图像处理的效果。给出型坡口多层多道焊边缘坐标的确定以及纠偏量和深度值的计算方法，实验说明效果较好。     

Automatic method of fruit object extraction under complex agricultural background for vision system of fruit picking robot

Fruit picking robot is required for agricultural automation for fruit harvest, and vision system is the important and crucial composition of a robot system. An automatic extraction method of fruit object under complex agricultural background for vision system in fruit picking robot is presented in this study. The method is based on an improving Otsu threshold algorithm using a new feature in OHTA color space. Color features are extracted in OHTA color space and then used as an input for the Otsu threshold algorithm which calculates the segmentation threshold value automatically. Four kinds of fruit images are selected to validate the automatic extraction method. The fruit objects are automatically extracted with this method and the outputs are presented in binary images. Numerous of experiments show that the automatic extraction method can extract mature fruit from complex agricultural background and the extraction accuracy is more than 95%. The results indicate an effective fruit object extraction method for vision system of fruit picking robot.     

一种基于单目视觉的移动机器人室内导航方法

针对室内导航的环境特点,提出了一种简单快速的、以踢脚线为参考目标的移动机器人室内导航方法。该方法从图像中提取踢脚线作为参考直线,通过两条直线在图像中的成像特征,提取角度和横向偏离距离作为移动机器人的状态控制输入,从而实现移动机器人的横向运动控制。该方法无需进行摄像机的外部参数标定,大大简化了计算过程,提高了视觉导航的实时性。     

Generalized dynamic model for multibodies manipulator

In this paper the general dynamical equations were given for multibodiesmanipulator.The system is a topologic tree structure consisting of arbitrary number ofrigid bodies.The hinges allow the rotational and/or tranlational motion.Inconsideration of influence of friction the dynamic equations are established by meansof Newton-Euler’s method.Further,the equations are separated by way ofconstructing the distribution matrices and a group of force and motion equations areobtained.     

Planning motion trace of robot by spinor method

ＩｎｔｒｏｄｕｃｔｉｏｎＴｒａｃｅｐｌａｎｎｉｎｇｉｓａｎｉｍｐｏｒｔａｎｔａｓｐｅｃｔｉｎｔｈｅｃｏｎｔｒｏｌｏｆｒｏｂｏｔ．Ａｃｃｏｒｄｉｎｇｔｏｔｈｅｄｅｍａｎｄｏｆｐｒｏｄｕｃｔｉｏｎ,ｐｏｓｉｔｉｏｎａｎｄａｔｔｉｔｕｄｅｔｒａｃｅｏｆｒｏ...     

对广义达朗贝尔运动方程的一点补遗

对传统的广义达朗贝尔运动方程作了两点推广：１）考虑含有移动关节的情形;２）把方程的适用范围由单链系统推广至树形系统。     

On Evolving Social Systems: Communication, Speciation and Symbiogenesis

In this paper we introduce three enhancements for evolutionary computing techniques in social environments. We describe the use of the genetic algorithm to evolve communicating rule-based systems, where each rule-based system represents an agent in a social/multi-agent environment. It is shown that the evolution of multiple cooperating agents can give improved performance over the evolution of an equivalent single agent, i.e. non-social, system. We examine the performance of two social system configurations as approaches to the control of gait in a wall climbing quadrupedal robot, where each leg of the quadruped is controlled by a communicating agent. We then introduce two social-level operators&2014;speciation and symbiogenesis&2014;which aim to reduce the amount of knowledge required a priori by automatically manipulating the system&2018;s social structure and describe their use in conjunction with the communicating rule-based systems. The reasons for implementing these kinds of operators are discussed and we then examine their performance in developing the controller of the wall-climbing quadruped. We find that the use of such operators can give improved performance over static population/agent configurations.     

Efficient generation of accurate mobility maps using machine learning algorithms

U.S Army’s mission is to develop, integrate, and sustain the right technology solutions for all manned and unmanned ground vehicles, and mobility is a key requirement for all ground vehicles. Mobility focuses on ground vehicles’ capabilities that enable them to be deployable worldwide, operationally mobile in all environments, and protected from symmetrical and asymmetrical threats. In order for military ground vehicles to operate in any combat zone, the planners require a mobility map that gives the maximum predicted speeds on these off-road terrains. In the past, empirical and semi-empirical techniques (Ahlvin and Haley, 1992; Haley et al., 1979) were used to predict vehicle mobility on off-road terrains such as the NATO Reference Mobility Model (NRMM). Because of its empirical nature, the NRMM method cannot be extrapolated to new vehicle designs containing advanced technologies, nor can it be applied to lightweight robotic vehicles.The mobility map is a function of different parameters such as terrain topology and profile, soil type (mud, snow, sand, etc.), vegetation, obstacles, weather conditions, and vehicle type and characteristics.A physics-based method such as the discrete element method (DEM) (Dasch et al., 2016) was identified by the NATO Next Generation NRMM Team as a potential high fidelity method to model the soil. This method allows the capture of the soil deformation as well as its non-linear behavior. Hence it allows the simulation of the vehicle on any off-road terrain and have an accurate mobility map generated. The drawback of the DEM method is the required simulation time. It takes several weeks to generate the mobility map because of the large number of soil particles (millions) even while utilizing high performance computing.One approach to reduce the computational time is to use machine learning algorithms to predict the mobility map. Machine learning (Boutell et al., 2004; Burges, 1998; Barber et al., 1997) can lead to very accurate mobility predictions over a wide range of terrains. Machine learning is divided into two categories: the supervised and the unsupervised learning. Supervised learning requires the training data to be labeled into predetermined classes, while the unsupervised learning does not require the training data to be labeled. Machine learning can help generate mobility maps using trained models created from a minimum number of simulation runs. In this study different supervised machine learning algorithms such as the support vector machine (SVM), the nearest neighbor classifier (k-NN), decision trees, and boosting methods were used to create trained models labeled as 2 classes for the ‘go/no-go’ map, 5 classes for the 5-speed map, and 7 classes for the 7-speed map. The trained models were created from the physics-based simulation runs of a nominal wheeled vehicle traversing on a cohesive soil.     

Review of terramechanics models and their applicability to real-time applications

ISTVS embarked on a project in 2016 that aims at updating the current ISTVS standards related to nomenclature, definitions, and measurement techniques for modelling, parameterizing, and, respectively, testing and validation of soft soil parameters and vehicle running gear-terrain interaction. As part of this project, a comprehensive literature review was conducted on the parameterization of fundamental terramechanics models. Soil parameters of the empirical models to assess off-road vehicle mobility, and parameters of the models to characterize the response of the terrain interacting with running gears or plates from the existing terramechanics literature and other researchers’ reports were identified. This review documents and summarizes the modelling approaches that may be applicable to real-time applications of terramechanics in simulation, as well as in controller design.