Lane change trajectory prediction by using recorded human driving data

Abstract

Being able to predict the trajectory of a human driver's potential lane change behavior in urban high way scenario is crucial for lane change risk assessment task. A good prediction of the driver's lane change trajectory makes it possible to evaluate the risk and warn the driver beforehand. Rather than generating such a trajectory only using a mathematical model, this paper develops a lane change trajectory prediction approach based on real human driving data stored in a database. In real-time, the system generates parametric trajectories by interpolating k human lane change trajectory instances from the pre-collected database that are similar to the current driving situation. In order to build this real lane change database, a human lane change data collection vehicle platform is developed. Extensive experiments have been carried out in urban highway environments to build a significant database with more than 200 lane changes. Real results show that this approach produces lane change trajectories that are quite similar to real ones which makes it suitable to predict humanlike lane change maneuvers.

Design and implementation of human driving data–based active lane change control for autonomous vehicles

“…The parameters of S D LK are determined by previous research based on driving data. 7 The above driving data are employed for deciding parameters of S D LC . From the driving data, c LC could be set to 10 m in Figure 3(a).…”

Attention-based Lane Change Prediction

“…Lane change prediction, being a fundamental building block for any autonomous driving task, is a hot topic in research and has been investigated for several years [6], [7], [8], [9], [10]. Picking the most informative features according to a criterion and then using "classical" methods, like SVMs or Random Forests [2], [3], [11], [12] contributed to the core of research in lane change prediction.…”

Lateral Motion Prediction of On-Road Preceding Vehicles: A Data-Driven Approach

“…We are particularly interested in the real-time on-road vehicle lateral motion prediction. Although significant efforts have been made by building mathematical models and conducting laboratory experiments [14,15,16], the actual environment on the road is much more complicated, which creates remarkable differences. Therefore, lab simulations can provide limited references for on-road motions.…”

“…Yao et al, in 2013, developed a parametric lane change trajectory prediction approach based on real human lane change data. This method generated a similar parametric trajectory according to the k-Nearest real lane change instances [15]. Kumar et al, in 2013, proposed an online learning-based approach to predict lane change intention, which incorporated support vector machine (SVM) and Bayesian filtering [16].…”