Object segmentation methods for online model acquisition to guide robotic grasping

A vision system is an integral component of many autonomous robots. It enables the robot to perform essential tasks such as mapping, localization, or path planning. A vision system also assists with guiding the robot's grasping and manipulation tasks. As an increased demand is placed on service robots to operate in uncontrolled environments, advanced vision systems must be created that can function effectively in visually complex and cluttered settings. This thesis presents the development of segmentation algorithms to assist in online model acquisition for guiding robotic manipulation tasks. Specifically, the focus is placed on localizing door handles to assist in robotic door opening, and on acquiring partial object models to guide robotic grasping. . First, a method for localizing a door handle of unknown geometry based on a proposed 3D segmentation method is presented. Following segmentation, localization is performed by fitting a simple box model to the segmented handle. The proposed method functions without requiring assumptions about the appearance of the handle or the door, and without a geometric model of the handle. Next, an object segmentation algorithm is developed, which combines multiple appearance (intensity and texture) and geometric (depth and curvature) cues. The algorithm is able to segment objects without utilizing any a priori appearance or geometric information in visually complex and cluttered environments. The segmentation method is based on the Conditional Random Fields (CRF) framework, and the graph cuts energy minimization technique. A simple and efficient method for initializing the proposed algorithm which overcomes graph cuts' reliance on user interaction is also developed. Finally, an improved segmentation algorithm is developed which incorporates a distance metric learning (DML) step as a means of weighing various appearance and geometric segmentation cues, allowing the method to better adapt to the available data. The improved method also models the distribution of 3D points in space as a distribution of algebraic distances from an ellipsoid fitted to the object, improving the method's ability to predict which points are likely to belong to the object or the background. Experimental validation of all methods is performed. Each method is evaluated in a realistic setting, utilizing scenarios of various complexities. Experimental results have demonstrated the effectiveness of the handle localization method, and the object segmentation methods.

Object segmentation methods for online model acquisition to guide robotic grasping

A vision system is an integral component of many autonomous robots. It enables the robot to perform essential tasks such as mapping, localization, or path planning. A vision system also assists with guiding the robot's grasping and manipulation tasks. As an increased demand is placed on service robots to operate in uncontrolled environments, advanced vision systems must be created that can function effectively in visually complex and cluttered settings. This thesis presents the development of segmentation algorithms to assist in online model acquisition for guiding robotic manipulation tasks. Specifically, the focus is placed on localizing door handles to assist in robotic door opening, and on acquiring partial object models to guide robotic grasping. . First, a method for localizing a door handle of unknown geometry based on a proposed 3D segmentation method is presented. Following segmentation, localization is performed by fitting a simple box model to the segmented handle. The proposed method functions without requiring assumptions about the appearance of the handle or the door, and without a geometric model of the handle. Next, an object segmentation algorithm is developed, which combines multiple appearance (intensity and texture) and geometric (depth and curvature) cues. The algorithm is able to segment objects without utilizing any a priori appearance or geometric information in visually complex and cluttered environments. The segmentation method is based on the Conditional Random Fields (CRF) framework, and the graph cuts energy minimization technique. A simple and efficient method for initializing the proposed algorithm which overcomes graph cuts' reliance on user interaction is also developed. Finally, an improved segmentation algorithm is developed which incorporates a distance metric learning (DML) step as a means of weighing various appearance and geometric segmentation cues, allowing the method to better adapt to the available data. The improved method also models the distribution of 3D points in space as a distribution of algebraic distances from an ellipsoid fitted to the object, improving the method's ability to predict which points are likely to belong to the object or the background. Experimental validation of all methods is performed. Each method is evaluated in a realistic setting, utilizing scenarios of various complexities. Experimental results have demonstrated the effectiveness of the handle localization method, and the object segmentation methods.

Introduction to this special section: Seismic imaging below complex overburdens

Despite decades of technology development and experience, seismic imaging below complex overburdens, such as salt bodies, basalt flows, or shallow gas accumulations, remains a challenge. For successful imaging in such settings, a number of key elements need to be in place: the overburden complexity needs to be properly captured and represented in the subsurface model (acquisition/model building), the target below the complex overburden needs to be sufficiently well illuminated (acquisition), and, finally, the target needs to be properly imaged through the complex overburden (imaging algorithm). All of these elements are discussed in the papers collected in this special section, which consists of contributions that demonstrate recent technology developments as well as insightful case studies that show how the different elements come together to make a successful imaging project.

Primary competencies for secondary school: psychological features of successful learning at transition grades

As the transition period from primary to secondary education is difficult for many students, our strive is to study the psychological features, which may scaffold student’s promotion and provide for future successful learning. We consider the educational media to be a complex system of psychological features, which depend on the content of education and the way, in which the educational process is organized. The “activity-and-content oriented” educational media devised after theoretical principles of the Developmental Instruction (Davydov), according to our hypothesis, provides sound bases for primary competencies formation, which are vital for secondary education. To test this assumption, we have diagnosed the quality of reflection, analysis, planning, model-acquisition and model-application, achieved by 204 students from two different educational media (“activity-and-content oriented” and “knowledge oriented”). For assessment purposes two diagnostic tools were used: “Transpositions” and “Moon” test. Both tests exploited contexts, which were unfamiliar to both groups of students. The results show, that students from the “activity-and-content” oriented educational media performed significantly better in most cognitive primary concepts. We consider these findings to support our major hypothesis, that educational media is a powerful source to scaffold students’ primary competencies’ formation, which in its turn provides for successful learning in secondary school.

MAi: An Intelligent Model Acquisition Interface for Interactive Specification of Dialogue Agents

The state of the art in automated conversational agents for enterprise (e.g. for customer support) require a lengthy design process with experts in the loop who have to figure out and specify complex conversation patterns. This demonstration looks at a prototype interface that aims to bring down the expertise required to design such agents as well as the time taken to do so. Specifically, we will focus on how a metawriter can assist the domain-writer during the design process and how complex conversation patterns can be derived from simplifying abstractions at the interface level.