Seongyong Koo

Telepresence robot system for English tutoring

This paper introduced a telepresence robot system for English tutoring. The system consists of the user system, the server system and the robot system. The robot is controlled by the collision prevention system, direct interface, server interface with assistant teacher and remote interface with native teacher. A remote native teacher had given lessons in English using this system for a month and several issues of telepresence robot system for remote lecture are found from the lecture.

Recognizing human intentional actions from the relative movements between human and robot

Human intention recognition is one of the most important research areas in the human-robot interaction in order to accomplish more natural and intelligent interaction. In the public service area, the basic four human intentional actions, ‘approach’, ‘depart’, ‘bypass’, and ‘stop’ were investigated by the humans movement pattern. In order to recognize those four human intentional actions, the mobile robot measured 360 degree distance between a robot and a human by two IR-scanner sensors, and three problems were solved: 1) human detection and tracking by improved K-means clustering method, 2) absolute human velocity estimation by Extended Kalman Filter, and 3) inferring intentional actions by HMM with position-dependent observation model. The result showed as quite good performance that inferred the probabilities of those four intentional actions.

Online touch behavior recognition of hard-cover robot using temporal decision tree classifier

Touch is obviously an important channel along with vision and speech for natural human robot Interaction. However, as most service robots are generally specialized for their own service, touch-centered shape design and additional costs/computation less related to their own tasks can represent a limit to the application of a touch system on a service robot. This paper originated from the motivation to apply a touch system with lower costs/computation to robots without design modifications. The proposed touch recognition system features hardware that is simply composed of charge-transfer touch sensor arrays, an accelerometer and a temporal decision tree classifier intended for online recognition and computational time reduction. Experiments performed by 12 people shows the practicability of the system. The results showed an average recognition rate of 83% with respect to the 4 touch patterns of hit, beat, rub and push.

A robust online touch pattern recognition for dynamic human-robot interaction

This paper presents a novel touch pattern recognition algorithm for dynamic proximate interaction between a robot and a human. At first, in order to guarantee reactive responses to various touch patterns, an online touch pattern algorithm is proposed based on a Temporal Decision Tree(TDT). Second, dynamic movements of a robot in a real interaction situation usually deteriorate the confidence level of the pattern classifier. A robust method to compensate for inconsistent recognition results in the dynamic interaction is proposed by a Consistency Index(CI), which estimates consistency degrees of human touch patterns over time. The algorithms are applied to a hard-cover touch recognition module, which is being developed for recognizing the four kinds of emotional touch patterns mainly used in human-robot affective interaction. The recognition performance is evaluated in a simple game scenario environment with KaMERo (KAIST Motion Expressive Robot), which is an emotionally interactive robot platform. The results show that the proposed algorithm guarantees commercially applicable recognition performance by compensating for the misclassification inherent in the dynamic movements of a robot.

Incremental object learning and robust tracking of multiple objects from RGB-D point set data

A novel approach for tracking multiple objects from RGB-D point set data.The incremental learning method allows tracking objects without prior knowledge.The robustness of the method was quantitatively analyzed in the interaction cases.The trade-off between the efficiency and accuracy was empirically examined.The limitations of the novel method were investigated in various moving objects. In this paper, we propose a novel model-free approach for tracking multiple objects from RGB-D point set data. This study aims to achieve the robust tracking of arbitrary objects against dynamic interaction cases in real-time. In order to represent an object without prior knowledge, the probability density of each object is represented by Gaussian mixture models (GMM) with a tempo-spatial topological graph (TSTG). A flexible object model is incrementally updated in the pro-posed tracking framework, where each RGB-D point is identified to be involved in each object at each time step. Furthermore, the proposed method allows the creation of robust temporal associations among multiple updated objects during split, complete occlusion, partial occlusion, and multiple contacts dynamic interaction cases. The performance of the method was examined in terms of the tracking accuracy and computational efficiency by various experiments, achieving over 97% accuracy with five frames per second computation time. The limitations of the method were also empirically investigated in terms of the size of the points and the movement speed of objects.

A dual-layer user model based cognitive system for user-adaptive service robots

This paper proposes a dual-layer user model to generate descriptive service recommendations for user-adaptive service robots. The user model represents user preferences as the associative memory in the bottom-layer and association rules in the top-layer. The learning and inference processes in the two layers, and the bottom-up rule extraction process, are explained. The proposed user model was applied to a user-adaptive coffee menu recommendation system, and the quantitative and qualitative performances of the user-adaptive and descriptive recommendation system were evaluated by comparison with non-descriptive and random recommendation methods.

Unsupervised object individuation from RGB-D image sequences

In this paper, we propose a novel unified framework for unsupervised object individuation from RGB-D image sequences. The proposed framework integrates existing location-based and feature-based object segmentation methods to achieve both computational efficiency and robustness in unstructured and dynamic situations. Based on the infants object indexing theory, the newly proposed ambiguity graph plays as a key component of the framework to detect falsely segmented objects and rectify them by using both location and feature information. In order to evaluate the proposed method, three table-top multiple object manipulation scenarios were performed: stacking, unstacking, and occluding tasks. The results showed that the proposed method is more robust than the location-only method and more efficient than the feature-only method.

GMM-based 3D object representation and robust tracking in unconstructed dynamic environments

Operating in unstructured dynamic human environments, it is desirable for a robot to identify dynamic objects and robustly track them without prior knowledge. This paper proposes a novel model-free approach for probabilistic representation and tracking of moving objects from 3D point set data based on Gaussian Mixture Model (GMM). GMM is inherently flexible such that represents any shape of objects as 3D probability distribution of the true positions. In order to achieve the robustness of the model, the proposed tracking method consists of GMM-based 3D registration, Gaussian Sum Filtering, and GMM simplification processes. The tracking performance of the proposed method was evaluated in the moving two human hands with one object, and it performed over 87% tracking accuracy together with processing 5 frames per second.

Multiple object tracking using an RGB-D camera by hierarchical spatiotemporal data association

In this paper, we propose a novel multiple object tracking method from RGB-D point set data by introducing the hierarchical spatiotemporal data association method (HSTA) in order to robustly track multiple objects without prior knowledge. HSTA is able to construct not only temporal associations between multiple objects, but also component-level spatiotemporal associations that allow the correction of falsely detected objects in the presence of various types of interaction among multiple objects. The proposed method was evaluated using the four representative interaction cases such as split, complete occlusion, partial occlusion, and multiple contacts. As a result, HSTA showed significantly more robust performance than did other temporal data association methods in the experiments.

Multiple people tracking from 2D depth data by deterministic spatiotemporal data association

This paper proposes a deterministic approach to track people in a populated environment from 2D depth data by a laser range finder attached on a mobile robot. This work aims to improve robustness of multiple people tracking in the presence of change of the number of people, missing data, and long-term occlusions by using spatiotemporal data association. The temporal data association method is based on the multi-frame tracking (MFT) and the improved MFT (IMFT) is proposed for enhancing computational efficiency in the long-term occlusions. A spatial data association algorithm used a matching algorithm from the leg history data for detecting a human subject from leg tracks. The proposed methodology has been assessed in the three walking patterns of two people and compared with MFT and MHT methods.