A self-learning lane change motion planning system considering the driver’s personality

Abstract

Nowadays, with more and more attention being paid to the characteristics and experience of drivers, a large number of driver classification algorithms have emerged. However, these methods basically cannot be adjusted independently to each driver. Therefore, this paper proposes a self-learning lane change motion planning system considering the driver’s personality. Firstly, the method of driver data acquisition and processing is determined to obtain and extract the lane change data. Then, the planning system built in this paper is explained from two aspects: lane change trigger and lane change trajectory. According to the artificial potential field theory, an obstacle driving risk field is established to evaluate the acceptance of environmental risks of different drivers, and to achieve personalized lane change triggers through online statistics. At the same time, the safety of lane change is ensured by establishing the safety distance model of the target lane. On the other hand, the driver characteristic coefficient Jc and the driver reaction and operation time td are introduced into the traditional Gaussian-distributed model to establish a personalized lane change trajectory planning model, in which the parameters are obtained from offline and online learning. Offline learning is based on DTW for trajectory matching, and uses AP clustering to obtain the generalized parameters; Online learning uses LSTM to achieve personalized updates. Finally, this paper selected 15 drivers for verification, and the results show that the motion planning system can well reproduce the lane change behavior of the driver.

Objective evaluation index for the comprehensive performance of intelligent vehicle lane-changing trajectory

“…By differentiating the lateral displacement in different orders, the expressions of lateral velocity, lateral acceleration and lateral jerk can be obtained, as shown in formulas (14) to (16). The stagnation point is obtained for lateral jerk, and the obtained moment of the lateral acceleration peak point is shown in formula (17). In addition, the expressions of lateral acceleration and lateral velocity peak are shown in formulas (18) and (19).…”

“…Previous research on intelligent vehicle trajectory planning control generally used trigonometric, 9,10 hyperbolic tangent, 11,12 polynomial, [13][14][15] Bessel, 16 Gaussian, 17 and artificial potential field 18 functions to generate lane-changing trajectories for intelligent vehicles. However, for the selection of the trajectory model, the goodness of fit between the model and the actual lane-changing trajectory was often adopted as the evaluation index in extant research.…”

A Review of Cockpit-driving Integration for Human-centric Autonomous Driving

“…The model achieved a prediction accuracy of 96.5% on the NGSIM dataset. Gao et al [68] utilized LC temporal information and historical LC data to extract personalized LC parameters through an LSTM network. These parameters were then applied to a Gaussian distribution equation to generate personalized LC trajectories.…”

A Review of Cockpit-driving Integration for Human-centric Autonomous Driving

“…The model achieved a prediction accuracy of 96.5% on the NGSIM dataset. Gao et al [68] utilized LC temporal information and historical LC data to extract personalized LC parameters through an LSTM network. These parameters were then applied to a Gaussian distribution equation to generate personalized LC trajectories.…”