Classification of Steps on Road Surface Using Acceleration Signals

Abstract

In order to reduce a road monitoring cost, we propose a system to monitor extensively road condition by cyclists with a smartphone. In this paper, we propose two methods towards road monitoring. First is to classify road signals to four road conditions. Second is to extract road signal from a smartphone's accelerometer in three positions: pants' side pocket, chest pocket and a bag in a front basket. In pants' side pocket, road signal is extracted by Independent Component Analysis. In chest pocket and bag in a front basket, road signal is extracted by selecting 1-axis affected from gravitational acceleration. In the experiment of the classification method, overall accuracy was 75%. The experimental results of the extraction methods with correlation coefficient showed the overall accuracy were more than 0.7 in pants' side pocket and chest pocket, the overall accuracy was less than 0.3 in bag in a front basket.

YKOB: Participatory Sensing‐Based Road Condition Monitoring Using Smartphones Worn by Cyclist

“…The participatory sensing via bicycles is endowed with a probe bicycle [19], which assesses road surface conditions and traveling environment, BikeNet [20] logging a cyclist's condition and environment data during bike riding, and sBike [21] capturing the contexts during the riding. [23], in contrast, we confirmed that the newly developed algorithm to classify the three types of road-step, positive step, negative step, and convex step, based on the waveform feature of the road surface signal was effective in discriminating between the artificial roadsteps and abnormal road-step. With the above background, we have been proposing YKOB (Your Kinetic Observation Bike) based on the premise of the participatory sensing, which inspects road surface conditions from the acceleration signal captured via smartphones held by the rider [22,23].…”

“…In Ref. [23], in contrast, we confirmed that the newly developed algorithm to classify the three types of road-step, positive step, negative step, and convex step, based on the waveform feature of the road surface signal was effective in discriminating between the artificial roadsteps and abnormal road-step. In this paper, the real mother wavelet (RMW) is augmented to enable discrimination of the four level differences including concave.…”

“…It has been clear in the preliminary experiment [23] that the acceleration signal captured at the rider's waist is most immune to the effect of bike rider's movement, and the road surface data emerge most intensely in the gravity direction component. It has been clear in the preliminary experiment [23] that the acceleration signal captured at the rider's waist is most immune to the effect of bike rider's movement, and the road surface data emerge most intensely in the gravity direction component.…”

“…A bicycle that is lighter in weight and slower in velocity than a car can detect even a very minor road step since very few bicycles are equipped with suspensions. With the above background, we have been proposing YKOB (Your Kinetic Observation Bike) based on the premise of the participatory sensing, which inspects road surface conditions from the acceleration signal captured via smartphones held by the rider [22,23]. The participatory sensing via bicycles is endowed with a probe bicycle [19], which assesses road surface conditions and traveling environment, BikeNet [20] logging a cyclist's condition and environment data during bike riding, and sBike [21] capturing the contexts during the riding.…”

“…Every one of the above methods, however, needs to mount some sensor for measurement on a bicycle that would make it difficult to secure many participants. With the above background, we have been proposing YKOB (Your Kinetic Observation Bike) based on the premise of the participatory sensing, which inspects road surface conditions from the acceleration signal captured via smartphones held by the rider [22,23]. The experiment in which a bike rider ran on the real road surface with a smartphone inserted in the trouser pocket [22], proved that the road surface signal can be extracted with high precision by applying the independent component analysis (ICA) for the issue of signal mixing of the observed signal with the signal of pedaling motion.…”