Abstract:We present here a general framework and a specific algorithm for predicting the destination, route, or more generally a pattern, of an ongoing journey, building on the recent work of [Y. Lassoued, J. Monteil, Y. Gu, G. Russo, R. Shorten, and M. Mevissen, "Hidden Markov model for route and destination prediction," in IEEE International Conference on Intelligent Transportation Systems, 2017]. In the presented framework, known journey patterns are modelled as stochastic processes, emitting the road segments visited during the journey, and the ongoing journey is predicted by updating the posterior probability of each journey pattern given the road segments visited so far. In this contribution, we use Markov chains as models for the journey patterns, and consider the prediction as final, once one of the posterior probabilities crosses a predefined threshold. Despite the simplicity of both, examples run on a synthetic dataset demonstrate high accuracy of the made predictions.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is **predicting the driver's travel pattern, destination or specific route**. Specifically, the author aims to establish a model that can predict the current journey in real - time by analyzing historical driving data. This not only helps to provide a personalized driving experience (such as personalized risk assessment, route planning, etc.), but can also be used to optimize problems such as the charging and discharging scheduling of hybrid vehicles. ### Problem Background 1. **Understanding Driver Intent**: Understanding the driver's destination and route is crucial for providing a personalized driving experience. For example, personalized risk assessment, route suggestions, and automatic route replanning can be carried out based on this. 2. **Predictability**: Research shows that 60% of trips can be predicted from the driving history, and more extensive research shows that more than 90% of user trajectories have potential predictability. Therefore, predicting the driver's travel pattern has important practical application value. ### Solution The paper proposes a method based on **Bayesian classifier and Markov chain** to predict the travel pattern. The main steps include: 1. **Modeling**: Model the known travel patterns as random processes, and these processes will send signals according to the road sections in the journey. 2. **Updating Posterior Probability**: As the journey progresses, continuously update the posterior probability of each travel pattern until the prediction stops when a certain posterior probability exceeds the preset threshold. 3. **Using Markov Chain**: Select Markov chain as the random process model because they are simple and easy to train and evaluate. ### Key Technical Points - **Bayesian Update**: Use Bayes' theorem to update the posterior probability of each travel pattern. - **Markov Chain**: Assume that the next road section in the journey depends only on the current road section, rather than the entire historical path. - **Likelihood Function**: Calculate the probability of observing the current sequence of road sections under a given travel pattern. ### Experimental Verification The author conducted experiments on synthetic data sets, and the results show that this method has high prediction accuracy and can make accurate predictions early in the journey. In conclusion, by combining the Bayesian classifier and Markov chain, this paper provides a flexible and effective framework for predicting the driver's travel pattern, thereby achieving a personalized driving experience and other related applications.

Bayesian Classifier for Route Prediction with Markov Chains

Predict Personal Continuous Route

Bus Alighting Stop Estimation Using Trip-chain and Bayesian Methodology.

A personal route prediction system based on trajectory data mining

A System for Destination and Future Route Prediction Based on Trajectory Mining

Bayesian Network for Red-Light-running Prediction at Signalized Intersections

Building Efficient Probability Transition Matrix Using Machine Learning from Big Data for Personalized Route Prediction

Destination Prediction by Trajectory Distribution Based Model

Modelling Changes in Travel Behaviour Mechanisms Through a High-Order Hidden Markov Model

Scalable Probabilistic Routes

Predicting Passenger's Public Transportation Travel Route Using Smart Card Data

Dynamic Prediction Of Drivers' Personal Routes Through Machine Learning

A Bayesian Network Approach To Time Series Forecasting Of Short-Term Traffic Flows

A bayesian network approach to traffic flow forecasting

Destination Prediction Based on Partial Trajectory Data

RouteKG: A knowledge graph-based framework for route prediction on road networks

Short-term traffic flow forecasting based on Markov chain model

Causal Conditional Hidden Markov Model for Multimodal Traffic Prediction

Train Delay Predictions Using Markov Chains Based on Process Time Deviations and Elastic State Boundaries

Generic Probabilistic Interactive Situation Recognition and Prediction: From Virtual to Real

Conditional forecasting of bus travel time and passenger occupancy with Bayesian Markov regime-switching vector autoregression