Takeshi Okadome

Object-based activity recognition with heterogeneous sensors on wrist

This paper describes how we recognize activities of daily living (ADLs) with our designed sensor device, which is equipped with heterogeneous sensors such as a camera, a microphone, and an accelerometer and attached to a users wrist. Specifically, capturing a space around the users hand by employing the camera on the wrist mounted device enables us to recognize ADLs that involve the manual use of objects such as making tea or coffee and watering plant. Existing wearable sensor devices equipped only with a microphone and an accelerometer cannot recognize these ADLs without object embedded sensors. We also propose an ADL recognition method that takes privacy issues into account because the camera and microphone can capture aspects of a users private life. We confirmed experimentally that the incorporation of a camera could significantly improve the accuracy of ADL recognition.

Generation of articulatory movements by using a kinematic triphone model

The method described here predicts the trajectories of articulatory movements for continuous speech by using a kinematic triphone model and the minimum-acceleration model. The kinematic triphone model, which is constructed from articulatory data obtained from experiments using an electro-magnetic articulographic system, is characterized by three kinematic features of a triphone and by the intervals between two successive phonemes in the triphone. After a kinematic feature of a phoneme in a given sentence is extracted, the minimum-acceleration trajectory that coincides with the extremum of the time integral of the squared magnitude of the articulator acceleration is formulated. The calculation of the minimum acceleration requires only linear computation. The method predicts both the qualitative features and the quantitative details of experimentally observed articulation.

WOZ experiments for understanding mutual adaptation

A robot that is easy to teach not only has to be able to adapt to humans but also has to be easily adaptable to. In order to develop a robot with mutual adaptation ability, we believe that it will be beneficial to first observe the mutual adaptation behaviors that occur in human–human communication. In this paper, we propose a human–human WOZ (Wizard-of-Oz) experiment setting that can help us to observe and understand how the mutual adaptation procedure occurs between human beings in nonverbal communication. By analyzing the experimental results, we obtained three important findings: alignment-based action, symbol-emergent learning, and environmental learning.

Object-Blog System for Environment-Generated Content

The object-blog service application automatically converts raw sensor data to environment-generated content (EGC), including texts, graphs, and figures. This conversion facilitates data searching and browsing. Generated content can serve several purposes, including memory aids, security, and communication media. In object-blog, personified objects automatically post entries to a Weblog about sensor data obtained from sensors attached to the objects. Feedback thus far from participants working with object-blog in an experimental environment has been positive.

Kinematic construction of the trajectory of sequential arm movements.

Abstract. A kinematic construction rule determining the trajectory of human sequential movements is formulated using minimum-jerk and minimum-angular-jerk trajectories. The kinematic construction rule states that the observed trajectory of sequential movements coincides with a weighted average of the minimum-jerk trajectory and the segmented minimum-angular-jerk trajectory. This rule covers not only point-to-point movements but also simple sequential movements. Five kinds of experiments that measure the trajectories in planar, multijoint sequential arm movements were conducted. The measured trajectories coincide with the predictions made on the basis of the kinematic construction rule presented here. Moreover, predictions of previous models such as the minimum-jerk, the equilibrium-trajectory, and the minimum-torque-change models are shown to be incompatible with our observations of sequential movements.

A Real-World Event Search System in Sensor Network Environments

Assuming an environment in which a sensor network always collects data produced by sensors attached to physical objects, the demonstration system presented here searches sensor data for corresponding real-world events using a natural language (NL) words in a query. The system translates each query into a physical quantity representation also introduced here, searches for a sensor data segment that satisfies the description by the representation, and returns the information about the event that is reflected in time series of values from sensor readings.

Formation conditions of mutual adaptation in human-agent collaborative interaction

When an adaptive agent works with a human user in a collaborative task, in order to enable flexible instructions to be issued by ordinary people, it is believed that a mutual adaptation phenomenon can enable the agent to handle flexible mapping relations between the human user’s instructions and the agent’s actions. To elucidate the conditions required to induce the mutual adaptation phenomenon, we designed an appropriate experimental environment called “WAITER” (Waiter Agent Interactive Training Experimental Restaurant) and conducted two experiments in this environment. The experimental results suggest that the proposed conditions can induce the mutual adaptation phenomenon.

Tweet Sentiment Analysis with Latent Dirichlet Allocation

The method proposed here analyzes the social sentiments from collected tweets that have at least 1 of 800 sentimental or emotional adjectives. By dealing with tweets posted in a half a day as an input document, the method uses Latent Dirichlet Allocation LDA to extract social sentiments, some of which coincide with our daily sentiments. The extracted sentiments, however, indicate lowered sensitivity to changes in time, which suggests that they are not suitable for predicting daily social or economic events. Using LDA for the representative 72 adjectives to which each of the 800 adjectives maps while preserving word frequencies permits us to obtain social sentiments that show improved sensitivity to changes in time. A regression model with autocorrelated errors in which the inputs are social sentiments obtained by analyzing the contracted adjectives predicts Dow Jones Industrial Average DJIA more precisely than autoregressive moving-average models.

Detection of social interaction from observation of daily living environments

In this article, we discuss how to detect occasional social interaction by a group of people in an open space such as a hall by observing the environment by cameras. Since it is known in the field of social psychology that some characteristic arrangement is maintained by each group of people during interaction, previous works have tried to detect social interaction based on the arrangement. However, these methods could confuse different groups especially when those groups are located close to each other, because the methods only consider direct relationship among the positions or orientations of the people for finding the characteristic arrangement. We propose a new region-based approach, which focuses on the spatial region to be occupied exclusively by each group of the people, introducing a technique for region extraction used in the field of image processing.

A Platform System for Developing a Collaborative Mutually Adaptive Agent

The characteristic task of service robots that can interact with humans is to achieve human-robot collaboration. Mutual adaptation is considered to be an important characteristic of robots, required for carrying out such collaborative tasks. Here, we introduce the concept of mutual adaptation, propose a learning model, and describe an experimental task to explain the above concept. A waiter robot performs a collaborative task using a platform system, which is developed by a constructive approach. The interactive and manual modes of this system are compared by performing a preliminary experiment to evaluate the effectiveness of the robots autonomous function. The results indicate that the robots autonomous function works well when operated in the interactive mode under short time or slow speed conditions.