GPS and Acceleration Data in Multimode Trip Data Recognition Based on Wavelet Transform Modulus Maximum Algorithm

Abstract

The GPS-based travel survey is an emerging data collection method in transportation planning. The survey's application in trip mode detection has been explored in many studies. Most research on trip mode detection methods based on GPS data has been developed and tested with data collected from European and American countries. The methods cannot be easily adapted to Asian countries such as China, India, and Japan, which have much higher population densities, more complex road networks, and highly mixed travel modes during daily commuting. Furthermore, for trip segment division in multimode travel, existing algorithms use travel time and distance thresholds that are highly dependent on local travel behavior and lack universality across traffic environments. This paper proposes an innovative framework for detecting trip modes in complex urban environments. First, a smartphone application, GPSurvey, was developed to collect passive GPS trace data. Then a wavelet transform modulus maximum algorithm was developed for trip segment division. The algorithm has outstanding capabilities for identifying singularity features of a signal; this factor suits the task of detecting mode changes in a complex traffic environment. A neural network module was developed for mode detection on the basis of cell phone GPS location and acceleration data. The results indicate that the proposed method has promising performance. The average absolute detection error of mode transfer time was within 1 min, and the accuracy for detecting all modes was greater than 85%.

CityTracks-RT: Uma aplicação para detecção do modo de transporte em tempo real nos centros urbanos

“…Em [15] foi utilizada a técnica de redes neurais para inferir o modo de transporte em cada segmento da viagem. Como resultado, foi detectado um nível de precisão na inferência de mais de 86%.…”

Travel Mode Detection Based on GPS Raw Data Collected by Smartphones: A Systematic Review of the Existing Methodologies

“…The reference lists of excluded reviews were reviewed and potential papers were gathered. Finally, 12 published papers matching all the criteria were included in this review [1,31,[38][39][40][41][42][43][44][45][46][47], as shown in Table 2. In particular, we selected [39] because the segment identification method of this study is a significant data preprocessing technique for travel mode detection.…”

“…However, the data preprocessing procedure was not mentioned in these papers [41,45]. The data preprocessing methods were simply referred to in papers [1,31,38,47].…”

“…Yang et al [41] adopted the wavelet transform modulus maximum (WTMM) algorithm for SI processing. The Gaussian family [Gaus (n)] was selected because it had the best performance in mode transfer time detection compared with Haar and Daubechies family.…”

“…Yang et al [41] selected the neural network (NN) algorithm to determine the travel mode of each trip segment. As a result, the accuracy of travel mode detection is more than 86%.…”