Showing papers by "Hiroshi Ishida published in 2006"

Mobile robot navigation using vision and olfaction to search for a gas/odor source

Abstract: This paper presents a new approach to search for a gas/odor source using an autonomous mobile robot. The robot is equipped with a CMOS camera, gas sensors, and airflow sensors. When no gas is present, the robot looks for a salient object in the camera image. The robot approaches any object found in the field of view, and checks it with the gas sensors to see if the object is releasing gas. On the other hand, if the robot detects the presence of gas while wandering around the area, it turns toward the direction of the wind that carries the gas. The robot then looks for any visible object in that direction. These navigation strategies are implemented into the robot under the framework of the behavior-based subsumption architecture. Experimental results on the search for a leaking bottle in an indoor environment are presented to demonstrate the validity of the navigation strategies.

Human-Inspired Robots

Abstract: Robots are just now becoming part of our everyday life and being used by ordinary people. Developing a humanlike appearance must go hand in hand with increasing robots' cognitive, social, and perceptive capabilities. In this installment of Trends & Controversies, Hiroshi Ishiguro and Minoru Asada present the ultimate imitation of a human's appearance: an android. Humanlike robots must communicate clearly using natural language. Stuart Shapiro outlines natural language's importance both as a practical way to communicate and a means of encouraging social interaction. Michael Thielscher presents the field of cognitive robotics, which investigates how robots can achieve greater autonomy and flexibility in such environments. Cynthia Breazeal has been a pioneer in developing socially interactive robots, and she presents the challenges and progress in this field. The growing field of socially assistive robotics is becoming an important testbed for human-inspired robots. Maja J. Mataric outlines the challenges of aiding people by means of social interaction rather than through physical contact alone. Finally, Hiroshi Ishida presents olfaction as a new sense for robots. The ability to recognize smells bring robots closer to humans

Indicators of Gas Source Proximity using Metal Oxide Sensors in a Turbulent Environment

Abstract: This paper addresses the problem of estimating proximity to a gas source using concentration measurements. In particular, we consider the problem of gas source declaration by a mobile robot equipped with metal oxide sensors in a turbulent indoor environment. While previous work has shown that machine learning classifiers can be trained to detect close proximity to a gas source, it is difficult to interpret the learned models. This paper investigates possible underlying indicators of gas source proximity, comparing three different statistics derived from the sensor measurements of the robot. A correlation analysis of 1056 trials showed that response variance (measured as standard deviation) was a better indicator than average response. An improved result was obtained when the standard deviation was normalized to the average response for each trial, a strategy that also reduces calibration problems

Estimation of Gas-Source Location Using Gas Sensors and Ultrasonic Anemometer

Abstract: This paper reports on a new method to search for a gas source. In the previous works on gas-source localization, robotic or human searchers were navigated using gas sensors and anemometers in order to track gas plumes and find their source locations. Instead, the proposed method provides a map showing the certainty of the gas-source location. In the proposed method, the area under consideration is meshed into a two dimensional grid of cells. At each observation of the gas concentration, the certainty of having a gas source with a specific strength is calculated for each cell using a turbulent diffusion model of gas distribution. The results are aggregated after repeating the observations at multiple points. The gas source is estimated to be in the cell for which high certainty values are consistently obtained. Successful demonstrations on estimating the gas-source location are presented to show the validity of the proposed method.

Robotic System for Localizing a Chemical Source Underwater by Mimicking Crayfish Behavior

Abstract: An underwater robotic system with a chemical sensing capability is proposed. Crayfish that prefer stagnant flow conditions are known to generate water currents by beating exopodites of maxillipeds to enhance the chemical perception. The developed robotic system consists of an array of electrochemical sensors and a pair of fanning devices mimicking the maxillipeds. The fanning devices beaten vertically enable the robot to draw water samples from a wide horizontal angular range and to detect the chemicals in the surroundings. Successful demonstrations are presented to show the robot's ability to search for a chemical source in a stagnant water pool.

Analysis of QCM gas sensor transient response by visualizing gas concentration

Abstract: Gas-source localization systems are required to quickly detect gas concentrations at various locations while moving. It is therefore necessary to design the system by taking account of the delay in the response and recovery of the gas sensor in the presence of variations in gas concentration. Also, in order to model the dynamic characteristics of the gas sensor in open air, it is essential to understand the delay in response and recovery of the gas sensor. To carry out a transient response analysis of such a gas sensor, it is necessary to know the gas concentration (changing in time) at the gas sensor. In the present study, in order to analyze the transient response of a QCM (quartz crystal microbalance) gas sensor, the gas concentration is visualized. A photodiode is placed close to the gas sensor and the intensity of scattered laser light is measured optically in order to know the concentration of the gas passing over the sensor. The sensor response and the intensity of scattered light are measured simultaneously and the sensor response is estimated from the visualized gas concentration by means of the models of a second-order system and a neural network. When the measured results and the estimated results are compared, the model based on the neural network is found to estimate more accurately. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 89(6): 14–21, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjb.10062