Automatic traffic modelling for creating digital twins to facilitate autonomous vehicle development

Abstract

A digital twin is often adopted in computer simulations to expedite autonomous vehicle developments by using the simulated 3D environment that reflects a physical environment. In particular, traffic simulations are a crucial part of training the driving logic before the field test of an autonomous vehicle is performed on specific regions to adapt to the region-specific, dynamic traffic conditions. Currently, the traffic conditions are either synthesised by tools (e.g. using mathematical models) or created manually (using domain knowledge), which cannot reflect the realistic, region-specific conditions or will require extensive labour works. In this article, we propose an automatic methodology to model the real-world traffic conditions captured by the sensor data and to reproduce the modeled traffic in the digital twin. We have built the tools based on the methodology and use the KITTI dataset to validate the effectiveness of the tools. To recreate the region-specific traffic, we present the results of capturing, modelling, and recreating the two-wheeler traffic condition on the Southeast Asia road. Our experimental results show that the proposed method facilitates the simulation of real-world, Southeast Asia-specific traffic conditions by removing the needs of the synthesised traffic and the labour hours.

Keywords:

• Traff modelling
• Digital twins
• Autonomous vehicle
• 3D simulations
• Modelling moving tracks

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We leverage the code of the modules to fit our needs, including lidar_euclidean_cluster_detect, native-L-shape fitting, and imm-ukf-pda tracker, which provide good performance (after our analyses) and avoid to reinvent the wheel.

2 For example, the lidar_euclidean_cluster_detect module is used to process the LiDAR data.

3 For instance, vehicle_sender is used to pass the commands to the simulated car, whereas mqtt_bridge_node is called to control the physical car.

4 The data is the city category, and the corresponding file name is 2011_09_29_drive_0071.

5 The green lines are orthogonal to the ego vehicle.

6 It is a dangerous, but common situation for a two-wheeler to cut in front of the vehicle. The coordinate of the experiment is at $\mathrm{N}22{}^{\circ }{55}^{\prime }{44.1}^{″}$ $\mathrm{E}120{}^{\circ }{16}^{\prime }{38.3}^{″}$

7 The demonstration video of the physical and simulated environments is available via the link: https://www.youtube.com/watch?v=2fKgzrVxzLo

Additional information

Funding

This work is financially supported in part by the ‘Intelligent Manufacturing Research Center’ (iMRC) from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan. This work is also supported by MOE, Taiwan, under the Manpower Cultivation Program of Autonomous Vehicles. This work is supported in part by the Ministry of Science and Technology, Taiwan [grant number MOST-110-2221-E-006-052], [grant number MOST-110-2218-E-006-026].