Online Map Matching Research Articles

AMM: An Adaptive Online Map Matching Algorithm

Online map matching is essential for some location-based services, such as car navigation. However, due to GPS measurement errors and/or the lack of sufficient information, the performance of most existing online algorithms will degrade in the increasingly complex traffic environment. In this paper, we propose an adaptive online map matching algorithm called AMM. The basic idea is that AMM should be able to calibrate GPS observation data for various measurement errors and under complex urban conditions. First, we establish a collaborative evaluation model between GPS points and candidate points to effectively filter low-quality GPS measurement points, dynamically set the weights of different features and comprehensively select the best candidate points. Second, we propose a retrospective correction mechanism to correct the previous matching results when more information is available, which will help improve the accuracy of future GPS points. Furthermore, we define parameter self-tuning rules for AMM to enhance its portability by avoiding time-consuming parameter tuning steps. We conduct extensive experiments to evaluate the performance of AMM on real vehicle trajectory datasets. The experiment results show that AMM outperforms its counterparts by up to 32% in terms of accuracy and its performance in different traffic conditions is more stable.

A robust map matching method by considering memorized multiple matching candidates

Map matching is to track the positions of vehicles on the road network based on the positions provided by GPS (Global Positioning System) devices. Balancing localization accuracy and computation efficiency is a key problem in map matching. Existing methods mainly use Hidden Markov Model (HMM) or historical transportation data to learn the transitional probabilities among road segments. Although the roads to explore can be remarkably reduced by the Markov assumption, miss-of-match and matching breaks may occur if the GPS data is highly noisy, and the transitional model needs to be learned offline. To address these problems, this paper presents Multiple Candidate Matching (MCM) to improve the robustness of map matching. MCM doesn't need to pre-train the transitional model nor the historical transportation information. MCM memorizes multiple historical matching candidates in the map matching process. It votes among historical matchings and current matchings, but generates limited number of road candidates in real-time to restrict the computation complexity. MCM for both online map matching and offline map matching are presented and their properties are analyzed theoretically and experimentally. Numerical experiments in large-scale data sets show that MCM is very promising in terms of accuracy, computational efficiency, and robustness. The matching break and miss-of-match problems can be resolved effectively when compared with the state-of-the-art map matching methods. Codes are outsourced at https://github.com/lindalee-inlab/MCM.

Online Trajectory Estimation Based on a Network-Wide Cellular Fingerprint Map.

Cellular signaling data is widely available in mobile communications and contains abundant movement sensing information of individual travelers. Using cellular signaling data to estimate the trajectories of mobile users can benefit many location-based applications, including infectious disease tracing and screening, network flow sensing, traffic scheduling, etc. However, conventional methods rely too much on heuristic hypotheses or hardware-dependent network fingerprinting approaches. To address the above issues, NF-Track (Network-wide Fingerprinting based Tracking) is proposed to realize accurate online map-matching of cellular location sequences. In particular, neither prior assumptions such as arterial preference and less-turn preference or extra hardware-relevant parameters such as RSS and SNR are required for the proposed framework. Therefore, it has a strong generalization ability to be flexibly deployed in the cloud computing environment of telecom operators. In this architecture, a novel segment-granularity fingerprint map is put forward to provide sufficient prior knowledge. Then, a real-time trajectory estimation process is developed for precise positioning and tracking. In our experiments implemented on the urban road network, NF-Track can achieve a recall rate of 91.68% and a precision rate of 90.35% in sophisticated traffic scenes, which are superior to the state-of-the-art model-based unsupervised learning approaches.

A Self-adjusting Online Map Matching Method

With the rapid development of intelligent transportation systems (ITS) and location-based services (LBS), it is more important to match the original trajectory sequence generated by users/vehicles to the actual road network. Most of the existing online map matching algorithms are based on the idea of local processing, or require richer data input and more mathematical models to ensure matching accuracy. This paper presents a simple and effective map matching method, called self-adjusting online map matching (AOMM). The algorithm is developed based on hidden Markov model (HMM). Considering the topological and geometric properties of the road network, the emission probability and transition probability calculation formulas of HMM are defined. And three adjustment strategies are provided to deal with trajectory noise points, dense trajectory points, and offset trajectory points. The algorithm only needs latitude and longitude information of trajectory points, and can match point by point. Experimental results on open trajectory data show that the algorithm has high matching accuracy and low output delay, and can meet the requirements of general online map matching tasks.

An Online Map Matching Algorithm Based on Second-Order Hidden Markov Model

Map matching is a key preprocess of trajectory data which recently have become a major data source for various transport applications and location-based services. In this paper, an online map matching algorithm based on the second-order hidden Markov model (HMM) is proposed for processing trajectory data in complex urban road networks such as parallel road segments and various road intersections. Several factors such as driver’s travel preference, network topology, road level, and vehicle heading are well considered. An extended Viterbi algorithm and a self-adaptive sliding window mechanism are adopted to solve the map matching problem efficiently. To demonstrate the effectiveness of the proposed algorithm, a case study is carried out using a massive taxi trajectory dataset in Nanjing, China. Case study results show that the accuracy of the proposed algorithm outperforms the baseline algorithm built on the first-order HMM in various testing experiments.

A Proposal of Online Map-matching Based Trajectory Compression Algorithm Using Road Networks

A Proposal of Online Map-matching Based Trajectory Compression Algorithm Using Road Networks

Incremental route inference from low-sampling GPS data: An opportunistic approach to online map matching

With the surging of smart device sensing and mobile networking, GPS data has been widely available for identifying vehicle position and route on the road map. For many real-time applications, such as traffic sensing and route recommendation, it is critical to immediately infer travelling route with incoming GPS data. In this paper, an opportunistic approach to online map matching is proposed to incrementally infer routes from low-sampling GPS data with low output latency. Unlike the hidden Markov model (HMM)-based approach, which often experiences certain delay between the GPS observation and inference, our algorithm can produce immediate inference when a new GPS point becomes available. Furthermore, a rollback mechanism is provided to correct the already inferred route when some abnormal situations are detected during the opportunistic inference process. We evaluate the proposed algorithm using real dataset of GPS trajectories over 100 cities around the world. Experimental results show that our algorithm is better than, or at least comparable to the state-of-the-art algorithms in terms of inference accuracy. More importantly, our algorithm can yield much shorter output latency and require less execution time, which is critical for many real-time navigation applications and location-based services.

Online Map Matching With Route Prediction

Map matching is a procedure that estimates the route traveled by vehicles or people by using observed coordinates. It is an important preprocessing procedure for location services based on global positioning system (GPS) data obtained from probe vehicles. One recently proposed major map matching approach is the hidden Markov model (HMM)-based method. However, HMM-based approaches suffer from latency, because they rely on the availability of future GPS points. This latency limits the ability of real-time traffic sensing and location services. This paper presents a novel online map matching algorithm that uses a probabilistic route prediction model instead of future GPS points. The probabilistic route prediction model can be trained by using historical trajectory data. Our experimental results show that the accuracy of the untrained proposed model is competitive with a naïve online HMM-based method without any latency. Moreover, the results show that the trained model obtains even higher accuracy. The experimental results also show that the proposed method is faster than the online HMM.

A Scalable Architecture for Real-Time Stream Processing of Spatiotemporal IoT Stream Data—Performance Analysis on the Example of Map Matching

Scalable real-time processing of large amounts of data has become a research topic of particular importance due to the continuously rising amount of data that is generated by devices equipped with sensing components. While existing approaches allow for fault-tolerant and scalable stream processing, we present a pipeline architecture that consists of well-known open source tools to specifically integrate spatiotemporal internet of things (IoT) data streams. In a case study, we utilize the architecture to tackle the online map matching problem, a pre-processing step for trajectory mining algorithms. Given the rising amount of vehicle location data that is generated on a daily basis, existing map matching algorithms have to be implemented in a distributed manner to be executable in a stream processing framework that provides scalability. We demonstrate how to implement state-of-the-art map matching algorithms in our distributed stream processing pipeline and analyze measured latencies.

A three-stage online map-matching algorithm by fully using vehicle heading direction

Map-matching is essential for almost all intelligent transportation systems, including context and personalized services. To support real-time intelligent transportation services, online map-matching is usually a prerequisite. Although many map-matching methods have been proposed, they often fail to balance the two conflicting objectives, i.e., matching quality and computation time. To alleviate the contradiction, in this paper, we propose a three-stage online map-matching algorithm, named as SD-Matching, to fully exploit a new dimension of collected GPS trajectory data (i.e., vehicle heading direction) in a provably smart way. In the first stage, heading direction is first used to enhance the probability computation of candidate edges for a given GPS point. In the second stage, heading direction is also employed to narrow down the searching space and serve as a cost-effective guider in the shortest path computation for two consecutive GPS points. In the third stage, heading direction is further utilized to refine the vehicle travelling path for a sequence of GPS points, together with the topology of the road network. Finally, we evaluate the SD-Matching algorithm using the real-world taxi data and road network data in the city of Beijing, China, to demonstrate its effectiveness and efficiency.

A Green Self-Adaptive Approach for Online Map Matching

GPS trajectory data plays a critical role for the intelligent transportation. Due to many factors like measure error, sampling error and battery power-saving requirement, the directly obtained trajectories will inaccurately align to the digital map. This requires an online map matching algorithm with high precision, low latency, and energy consumption saving. However, the existing approaches need to make some trade off among these criterions. Hence, based on the hidden Markov model, this paper proposes an adaptive online map matching algorithm to improve the performances in these perspectives at the same time: 1) the probabilistic method integrating the geometric information and topological information is developed to improve the accuracy; 2) the adaptive sampling frequency method is proposed to reduce the energy consumption; and 3) the adaptive sliding window method is presented to reduce the output delay. The experiments demonstrate that our approaches can not only improve the matching precision, but also reduce the latency and energy consumption simultaneously.

Unsupervised Part-Based Scene Modeling for Map Matching

In exploring the 1-to-Nmap matching problem that exploits a compact map data description, we hope to improve map matching scalability used in robot vision tasks. We propose explicitly targeting fast succinct map matching, which consists of map matching subtasks alone. These tasks include offline map matching attempts to find compact part-based scene models that effectively explain individual maps by using fewer larger parts. These tasks also include online map matching to find correspondence between part-based maps efficiently. Our part-based scene modeling approach is unsupervised and uses common pattern discovery (CPD) between input and known reference maps. Results of our experiments, which use a publicly available radish dataset, confirm the effectiveness of our proposed approach.

2A2-T01 部品SLAM : 教師無し部品モデルによる高速・簡潔な間接型マップマッチング(移動ロボットの自己位置推定と地図構築)

In this paper, we explore the challenging 1-to-N map matching problem, which exploits a compact description of map data, to improve the scalability of map matching techniques used by various robot vision tasks. We propose a first method explicitly aimed at fast succinct map matching, which consists only of map-matching subtasks. These tasks include offline map matching attempts to find a compact part-based scene model that effectively explains each map using fewer larger parts. The tasks also include an online map matching attempt to efficiently find correspondence between the part-based maps. Our part-based scene modeling approach is unsupervised and uses common pattern discovery (CPD) between the input and known reference maps. This enables a robot to learn a compact map model without human intervention. We also present a practical implementation that uses the state-of-the-art CPD technique of randomized visual phrases (RVP) with a compact bounding box (BB) based part descriptor, which consists of keypoint and descriptor BBs. The results of our challenging map-matching experiments, which use a publicly available radish dataset, show that the proposed approach achieves successful map matching with significant speedup and a compact description of map data that is tens of times more compact. Although this paper focuses on the standard 2D point-set map and the BB-based part representation, we believe our approach is sufficiently general to be applicable to a broad range of map formats, such as the 3D point cloud map, as well as to general bounding volumes and other compact part representations.