현재 진행되고 있는 대부분의 로봇 연구는 특정한 목적을 수행함에 있어서 기능적으로 안정적이고 원활한 움직임을 보이는 것과 더불어 사용자가 원하는 정보를 적절하게 전달하는 것에 초점을 맞춰 진행하고 있다. 본 논문에서는 로봇의 작업수행 상태가 아닌 외부로부터 입력 값이 없는 대기상태의 행동패턴을 제시하여 보다 자연스러운 인간-로봇 상호작용을 제시하고자 하였다. 로봇의 대기상태 행동패턴을 디자인함에 있어서 비디오 판독 방법을 선택하였고, 실제 서비스업에 종사하는 10명의 사람들을 비디오로 녹화하여 사람들과 상호작용이 없는 대기상태의 반복적인 행동패턴을 관찰하였다. 각각의 비디오 데이터로부터 총 21개의 반복적 행동을 기록하였고, 비슷한 항목들을 통합하여 최종적으로 11개의 행동패턴을 추출하였다. 추출된 패턴들 가운데 6개의 대표적인 행동들을 EEEX라는 로봇에 적합하도록 인코딩 작업을 하였고, 이것들이 사람들에게 올바르게 인식되는지를 확인하기 위하여 검증실험을 수행하였다. 사람들은 대부분의 로봇 행동패턴을 실험에서 의도한 바와 같이 인식하였으나 로봇의 하드웨어 특성상 몸과 팔의 움직임에서 약간의 혼동 요소가 있었다. 추후 실험을 통해 EEEX를 대형 마트의 입구에 실제로 배치하여 대기 중에 중립행동을 보일 때와 보이지 않을 때의 손님의 관심도 차이를 조사해보고자 한다.

This paper presents a vision-based fall detection system to automatically monitor and detect people’s fall accidents, particularly those of elderly people or patients. For video analysis, the system should be able to extract both spatial and temporal features so that the model captures appearance and motion information simultaneously. Our approach is based on 3-dimensional convolutional neural networks, which can learn spatiotemporal features. In addition, we adopts a thermal camera in order to handle several issues regarding usability, day and night surveillance and privacy concerns. We design a pan-tilt camera with two actuators to extend the range of view. Performance is evaluated on our thermal dataset: TCL Fall Detection Dataset. The proposed model achieves 90.2% average clip accuracy which is better than other approaches.

In this paper we propose a haptic interaction system that physically represents the underlying geometry of objects displayed in a 2D picture, i.e., a digital image. To obtain the object’s geometry displayed in the picture, we estimate the physical transformation between the object plane and the image plane based on homographic information. We then calculate the rotated surface normal vector of the object’s face and place it on the corresponding part in the 2D image. The purpose of this setup is to create a force that can be rendered along with the image without distorting the visual information. We evaluated the proposed haptic rendering system using a set of pictures of objects with different orientations. The experimental results show that the participants reliably identified the geometric configuration by touching the object in the picture. We conclude this paper with a set of applications.

This paper presents a new miniature haptic display to convey ample haptic information to a user of a handheld interface. There are buttons on interfaces or general electronic devices, but existing buttons provide haptic feedback of only one passive pattern to a user. Because humans perceive tactile and kinesthetic information simultaneously when they handle objects the proposed actuator provides both sensations at once. It is able to generate various levels of kinesthetic sensations when pressing a button under diverse situations. Also, vibrotactile feedback can be delivered for exciting haptic effects with numerous patterns. Its performance was evaluated in accordance with the resistive force by changing the intensity of the input current. Experiments show that the proposed actuator has the ability to provide numerous haptic sensations for more realistic and complex haptic experiences.

This research would investigate deeply the operation of an omni-directional mobile robot that is able to move with high acceleration. For the high acceleration performance, the vehicle utilizes the structure of Active Split Offset Casters (ASOCs). This paper is mainly focused on inverse kinematics of the structure, hardware design to secure durability and preserve the wheels’ contact to the ground during high acceleration, and localization for the real time position control.

Emotion interaction between human and robot is an important element for natural interaction especially for service robot. We propose a hybrid emotion generation architecture and detailed design of reactive process in the architecture based on insight about human emotion system. Reactive emotion generation is to increase task performance and believability of the service robot. Experiment result shows that it seems possible for the reactive process to function for those purposes, and reciprocal interaction between different layers is important for proper functioning of robot’s emotion generation system.

This paper presents a theoretical study for making intelligent robots with human-like mind. For the development of a cognitive mental model, we developed three algorithms based on the cognitive process for human psychoanalysis. Specifically, the concept of id, ego and superego from the theory of Sigmund Freud was adopted and the procedural algorithms were presented.

This paper introduces the research progress on the artificial brain in the Telerobotics and Control Laboratory at KAIST. This series of studies is based on the assumption that it will be possible to develop an artificial intelligence by copying the mechanisms of the animal brain. Two important brain mechanisms are considered: spike-timing dependent plasticity and dopaminergic plasticity. Each mechanism is implemented in two coding paradigms: spike-codes and rate-codes. Spike-timing dependent plasticity is essential for self-organization in the brain. Dopamine neurons deliver reward signals and modify the synaptic efficacies in order to maximize the predicted reward. This paper addresses how artificial intelligence can emerge by the synergy between self-organization and reinforcement learning. For implementation issues, the rate codes of the brain mechanisms are developed to calculate the neuron dynamics efficiently.