Kuya Takami

Non-Field-of-View Acoustic Target Estimation in Complex Indoor Environment

This paper presents a new approach which acoustically localizes a mobile target outside the Field-of-View (FOV), or the Non-Field-of-View (NFOV), of an optical sensor, and its implementation to complex indoor environments. In this approach, microphones are fixed sparsely in the indoor environment of concern. In a prior process, the Interaural Level Difference IID of observations acquired by each set of two microphones is derived for different sound target positions and stored as an acoustic cue. When a new sound is observed in the environment, a joint acoustic observation likelihood is derived by fusing likelihoods computed from the correlation of the IID of the new observation to the stored acoustic cues. The location of the NFOV target is finally estimated within the recursive Bayesian estimation framework. After the experimental parametric studies, the potential of the proposed approach for practical implementation has been demonstrated by the successful tracking of an elderly person needing health care service in a home environment.

Estimation of a nonvisible field-of-view mobile target incorporating optical and acoustic sensors

This paper presents a nonvisible field-of-view (NFOV) target estimation approach that incorporates optical and acoustic sensors. An optical sensor can accurately localize a target in its field-of-view whereas the acoustic sensor could estimate the target location over a much larger space, but only with limited accuracy. A recursive Bayesian estimation framework where observations of the optical and acoustic sensors are probabilistically treated and fused is proposed in this paper. A technique to construct the observation likelihood when two microphones are used as the acoustic sensor is also described. The proposed technique derives and stores the interaural level difference of observations from the two microphones for different target positions in advance and constructs the likelihood through correlation. A parametric study of the proposed acoustic sensing technique in a controlled test environment, and experiments with an NFOV target in an actual indoor environment are presented to demonstrate the capability of the proposed technique.

Bayesian non-field-of-view target estimation incorporating an acoustic sensor

This paper presents non-field-of-view (NFOV) target estimation incorporating an acoustic sensor, which consists of two microphones. The proposed approach derives the interaural level difference (ILD) of observations from the two microphones for different target positions and stores the ILDs as database a priori. Given a new acoustic observation on a target, an acoustic observation likelihood is created by calculating the correlation of the ILD of the new observation to the stored ILDs. A joint observation likelihood is then developed by fusing the optical and acoustic observation likelihoods, and the recursive Bayesian estimation updates and maintains belief on the target using the joint observation likelihood. The proposed approach detects a target positively using an acoustic sensor even if it is outside the field of view of the optical sensor and localizes the target accurately by estimating it within the RBE. The efficacy of the proposed approach was first validated by experimental studies. Further numerical demonstrations then show the applicability of the proposed approach to the NFOV target estimation.

Non-field-of-view indoor sound source localization based on reflection and diffraction

This paper presents a new acoustic approach to locate a mobile target outside the field-of-view (FOV), or the non-field-of-view (NFOV) of an optical sensor, based on the reflection and diffraction signals. In this approach, a sensor platform determines a reflection time-difference-of-arrival (TDOA) and frequency dependent diffraction to two distinct observation likelihoods from a single targets sound. The fusion of these likelihoods, a joint acoustic observation likelihood, estimate the NFOV target probabilistically within the recursive Bayesian estimation (RBE) framework. The approach was formulated and derived mathematically. Through parametric studies in simulation, the potential of the proposed approach for practical implementation has been demonstrated by the successful localization of the sound source. Finally, a preliminary validation of sound separation was performed in a controlled experimental environment showing the difference between diffraction alone and combination of diffraction and reflection signals.

Acoustic recursive Bayesian estimation for non-field-of-view targets

This paper presents acoustic recursive Bayesian estimation (RBE) for non-field-of-view (NFOV) targets. The proposed approach uses an acoustic sensor, which consists of two microphones. Given a new acoustic observation on a target, an acoustic observation likelihood is created, and the RBE updates and maintains the target belief using the acoustic observation likelihood. The proposed approach detects a target positively even if it is outside the field of view of the optical sensor and locates the target accurately by estimating it within the RBE. The effectiveness of the acoustic observation likelihood was first validated experimentally. Further experimental studies then demonstrated the applicability of the proposed RBE approach to the NFOV target estimation.

High-Resolution Deformation Measurement System for Fast Rotating Tires

This paper presents a high-resolution measurement system for fast-rotating tires. The developed system is compatible with and can be integrated with the tire testing machines used in industry research facilities. A field programmable gate array controller with a high clock rate triggers a short duration strobe flash capturing a clear instant image of a rotating tire based on an encoder reference signal. Since the rotation of a tire on a testing machine is periodic, the system can effectively capture the deformation of a rotating tire at a high equivalent sampling frequency. The Complementary Metal-Oxide-Semiconductor (CMOS)-based high-resolution low-cost system can be employed to measure tire deformation associated with sound generation. This was implemented by synchronously measuring sound with a microphone array. The validity of the developed system was investigated by experimental evaluation. Then, the system was implemented at the tire testing facility, and tested by rotating a tire at a drum speed up to 100 km/h and capturing images at every 0.025 degree rotation.Copyright

Prediction of Railroad Track Foundation Defects Using Wavelets

This paper presents a novel technique that utilizes wavelet analysis to identify and predict the defects in railroad foundations and rails to prevent derailment or other damages. The proposed defect detection algorithm eliminates the use of wheel and/or track monitoring systems, which are expensive and time inefficient. The algorithm has been validated for the rail crack prediction using only vertical accelerometer signal which accurately detects impending rail breakage while distinguishing the signal generated by special track components such as rail joins and switches. Since the algorithm is flexible, further development can be tailored to detect significantly different rail defects such as track shift and other rail foundation defects. The algorithm is further improved by incorporating SIMPACK dynamic simulation to assist classification of the acceleration signatures. The actual data was then compared to simulation in order to validate the effectiveness of the algorithm.© 2013 ASME

Non-field-of-view sound source localization using diffraction and reflection signals

This paper describes a non-field-of-view (NFOV) localization approach for a mobile robot in an unknown environment based on an acoustic signal combined with the geometrical information from an optical sensor. The approach estimates the location of a target through the mobile robots sensor observation frame, which consists of a combination of diffraction and reflection acoustic signals and a 3-D environment geometrical description. This fusion of audio-visual sensor observation likelihoods allows the robot to estimate the NFOV target. The diffraction and reflection observations from the microphone array generate the acoustic joint observation likelihood. The observed geometry also determines far-field or near-field acoustic conditions to improve the estimation of the sound direction of arrival. A mobile robot equipped with a microphone array and an RGB-D sensor was tested in a controlled environment, an anechoic chamber, to demonstrate the NFOV localization capabilities. This resulted in +/- 18 degrees, and less than 0.75 m error in angle and distance estimation, respectively.

Map-based navigation of an autonomous car using grid-based scan-to-map matching

This paper presents the map-based navigation of a car with autonomous capabilities using grid-based scan-to-map matching. The autonomous car used for demonstration is built based on Toyota Prius and can control the throttle, the brake and the steering by a computer. The proposed grid-based scan-to-map matching method represents a map with a finite number of grid cells, represents a scan and the map with scan points at each grid as normal distributions (NDs) and constructs a map by matching the scan NDs to the map NDs. The proposed method enables scan-based mapping at high speed while maintaining high accuracy. The representation of a grid cell of a map in terms of multiple NDs further enhances speed and accuracy. The accuracy analysis of the proposed method shows that a small robot with a wheel diameter of 8cm had yielded no loop closure error after the travel of 186m while the terminal position error by the GMapping was approximately 1m with the error growth of 1%. The application of the proposed method with the autonomous car has then demonstrated the ability of the proposed method for autonomous driving with varying and high speed and has also quantified the significance of speed for successful mapping in autonomous driving.Copyright