Jiwon Shin

Teaching a core CS concept through robotics

We implemented single-session workshops using the Thymio-II--a small, self-contained robot designed for young students, and VPL--a graphical software development environment based upon event handling. Our goal was to investigate if the students could learn this core computer science concept while enjoying themselves in the robotics context. A visual questionnaire was developed based upon the combined Bloom and SOLO taxonomies, although it proved difficult to construct a questionnaire appropriate for young students. We found that--despite the short duration of the workshop--all but the youngest students achieved the cognitive level of Unistructural Understanding, while some students achieved higher levels of Unistructural Applying. and Multistructural Understanding and Applying.

Unsupervised discovery of repetitive objects

We present a novel approach for unsupervised discovery of repetitive objects from 3D point clouds. Our method assumes that objects are non-deformable and uses multiple occurrences of an object as the evidence for its existence. We segment input range data by superpixel segmentation and extract features for each segment. We search for a group of segments where each segment matches a segment in another group using a joint compatibility test. The discovered objects are then verified by the Iterative Closest Point algorithm to remove false matches. The presented method was tested on real data of complex objects. The experiments demonstrate that the proposed approach is capable of finding objects that occur multiple times in a scene and distinguish apart those objects of different types.

Object classification based on a geometric grammar with a range camera

This paper proposes an object classification framework based on a geometric grammar aimed for mobile robotic applications. The paper first discusses the geometric grammar as a compact representation form for object categories with primitive parts as its constituent elements. The paper then discusses the object classification implemented as parsing of primitive parts. In particular, two approaches are discussed that constrain the search space in order to render the parsing of the primitive parts practical. The two approaches are experimentally verified, first, for a generic object category of chair applied to real range images acquired with a range camera mounted on a mobile robot and, second, for multiple generic object categories applied to synthetic range images. The experimental results show the practicability of the framework.

Automatic speed control for SmartWalker

Ensuring mobility of the elderly is an important task in our aging society. To this end, this paper presents an automatic speed controller for the SmartWalker -- a high-tech extension of a regular walker. The walker locates its user by detecting the users legs using a laser range scanner. The controller then determines the optimal speed for the walker using the users location and other sensory data. We evaluated the walker and its speed controller with thirteen residents at three different retirement homes. Our analysis showed that the walker with the controller is slightly more comfortable and easier to maneuver than the walker without the controller and is more liked than traditional walkers.

Concurrency patterns for easier robotic coordination

Software design patterns are reusable solutions to commonly occurring problems in software development. Growing complexity of robotics software increases the importance of applying proper software engineering principles and methods such as design patterns to robotics. Concurrency design patterns are particularly interesting to robotics because robots often have many components that can operate in parallel. However, there has not yet been any established set of reusable concurrency design patterns for robotics. For this purpose, we choose six known concurrency patterns - Future, Periodic timer, Invoke later, Active object, Cooperative cancellation, and Guarded suspension. We demonstrate how these patterns could be used for solving common robotic coordination tasks. We also discuss advantages and disadvantages of the patterns and how existing robotics programming frameworks can support them.

SmartWalker: Towards an Intelligent Robotic Walker for the Elderly

This paper presents SmartWalker and evaluates the appropriateness and usefulness of the walker and its gesture-based interface for the elderly. As a high-tech extension of a regular walker, the SmartWalker aims to assist its user intelligently and navigate around its environment autonomously. Equipped with sensors and actuators, the prototype accepts gesture commands and navigates around accordingly. The gesture-based interface uses a k-nearest neighbours classifier with dynamic time warping to recognize gestures and the Viola and Jones face detector to locate the user. We evaluated the walker with 23 residents and eight staff members at five different retirement homes in Zurich. The elderly found the SmartWalker useful and exciting, but few were willing to replace their walkers by robotic walkers. Their reluctance may stem from the walkers size and weight and their unfamiliarity with technology.

SmartWalker: An intelligent robotic walker

Ensuring mobility of the elderly is an important task in our aging society. To this end, this paper presents SMARTWALKER, a high-tech extension of a regular walker that aims to navigate around its environment autonomously and assist its user intelligently. The walker is equipped with sensors and actuators and operates in two modes, autonomous and assistive. In the autonomous mode, the walker accepts gesture commands via its gesture-based interface and navigates around accordingly. The interface uses a k-nearest neighbors classifier with dynamic time warping to recognize gestures and the Viola and Jones face detector to locate the user. In the assistive mode, its automatic speed controller determines the optimal speed for the walker. The walker locates its user by detecting the user’s legs using a laser range scanner and combines the information with other sensory data for the speed control. The walker was evaluated with residents of retirement homes in Zürich. The study aimed to understand the usefulness of robotic walker to the elderly, evaluate its potential to replace traditional walkers, and determine the appropriateness of the added functionality. The elderly found the SMARTWALKER’s autonomy useful and exciting and the walker with the controller slightly more comfortable and easier to maneuver. They stated that the walker is too big and too heavy but liked it more than traditional walkers in the assistive mode.