**1. Introduction**

During the operation of the road, the road surface will inevitably suffer from some defects or damage due to the crushing, impact, and weather changes of the passing vehicles. These defects and damage are often referred to as abnormal road conditions [1]. Abnormal road conditions have a negative impact on vehicle speed, fuel consumption, mechanical wear, ride comfort, and even safety. The traditional abnormal road condition information collection mainly relies on the manual site survey and special patrol vehicle, which is inefficient and high in cost. According to the statistics of the Ministry of Transport of China, the maintenance expenditure of the national toll road in 2017 has reached 53.39 billion yuan.

In order to save money and time costs, many experts and scholars have studied abnormal road surface identification methods [2,3]. There are three main methods for identifying abnormal roads: Visual [4], three-dimensional reconstruction [5,6], and dynamic vehicle response. The method of road surface recognition based on visual or three-dimensional reconstruction has high-performance requirements and high cost, which is not conducive to comprehensive promotion and use. The test scope is also very limited. The method based on the smartphone acceleration sensor to identify the abnormal road surface has certain advantages [2].

In early research, accelerometers are used for pavement recognition. De Zoysa [7] proposed to deploy a small number of mobile sensors in the public transportation system to detect road conditions, but the system exhibited low detection e fficiency and a false-positive rate. Chen [8] proposed a crowdsourcing based road surface monitoring system. By adding acceleration sensors and GPS modules to the vehicle to obtain the acceleration, velocity, and position information of the vehicle, the Gaussian background model is used to identify the abnormal road surface. Based on the research of Chen, Harikrishnan [9] improved the Gaussian background model and proposed an abnormal road surface recognition method that can adapt to di fferent vehicle speeds. This method could classify the speed bump and road surface bumps according to the X-Z axis acceleration ratio.

In order to improve the accuracy of road surface recognition, Wang [10] proposed a method by fusing the feature data from an acceleration sensor and camera to identify the abnormal road type. Celaya [11] installed sensors at the front of the vehicle to obtain the vehicle vibration response when the vehicle passed the speed bump and used the multivariate genetic algorithm to detect the road surface anomaly. This method can realize the recognition of abnormal road surfaces with a low false alarm rate, but the calculation is complicated, and a large number of statistical features such as mean, variance, peak, and standard deviation are needed for machine learning. With the rapid development of mobile intelligent terminal technology, smartphones equipped with sensors such as accelerometers and global positioning navigation systems can be used to detect abnormal road surfaces [12]. Cong [13] used the probabilistic statistical method and wavelet analysis method to establish an identification model of abnormal data and uses median filtering and wavelet filtering to process the data so that the data detected by the smartphone can truly reflect the vibration of the vehicle. Yi [14] proposed a smartphone detection vehicle to monitor the road surface, using an anomaly indexing algorithm to detect the speed bump, the pit, and the manhole cover. However, in the subsequent experiments, it was found that the proposed algorithm can only identify the speed bump, and the identification of other abnormal road conditions is not favorable. Mukherjee [15] established a quarter-vehicle model and a half-vehicle model and studied the acceleration response of the vehicle when crossing the deceleration belt. They developed a mathematical statistics method to identify the deceleration belt. Zhao [16] proposed a feature extraction method combining time-domain parameter characteristics and wavelet packet energy characteristics based on vehicle suspension vibration response and used a probabilistic neural network to classify road surface.

In summary, the research on the road surface condition identification method has achieved certain results. Recent studies have proven that smartphone accelerometers can e ffectively capture vehicle vibrations caused by abnormal road surfaces. By analyzing the signals from these mobile sensors, we have the potential to identify road anomalies. In this study, a Gaussian background model is used to identify abnormal roads, and an adaptive adjustment mechanism based on vehicle speed is proposed to improve the recognition accuracy. The parameters of the Gaussian background model are optimized by using fuzzy logic inference machines, making the method suitable for di fferent types of vehicles, and using the kNN algorithm to classify abnormal roads.

#### **2. Road Information Sharing System**

Obtaining abnormal road information in advance can e ffectively prevent tra ffic accidents, but due to road geometry, weather, lighting conditions, etc., the driver may not be able to notice the abnormal road ahead. The information-sharing technology is used to construct an abnormal road sharing system, especially encouraged by the development of the vehicle network and intelligent transportation system [17–21]. The road information sharing system is designed to promptly warn the driver when the driver approaches an abnormal road at a dangerous speed. In addition, the relevant information of the abnormal road surface is sent to the municipal road maintenance unit, so that the road maintenance personnel can timely ge<sup>t</sup> the road damage status and repair it to ensure the safety and comfort of travel [22–25].

In order to monitor the road surface conditions in realtime, it is necessary to find an efficient and reliable communication technology to transmit abnormal road information. This paper uses existing cellular network technology to collect abnormal road surface data. Figure 1 shows an architectural diagram of a road information sharing system. A smartphone with an acceleration sensor and a GPS module is fixed on the vehicle to obtain the latitude and longitude coordinates of the vehicle, the traveling speed, and the vibration acceleration data of the vehicle body. When the vehicle passes the abnormal road surface, the smartphone will upload the location and type of the abnormal road surface to the cloud. The abnormal road information is sent to the road maintenance personnel. When other vehicles approach the abnormal road surface, the cloud will issue an abnormal road surface reminder to ensure that the vehicle can pass the area safely and smoothly.

**Figure 1.** Road information sharing system architecture diagram.
