Reinforcement Learning - Train Agent for Three-Lane Highway

5 次查看（过去 30 天）
Christian Davila
Christian Davila 2021-10-4
回答： Salman Ahmed 2021-11-17
Hello everyone,
I am trying to train an agent for a scenario below:
An autonomous car in a three-lane highway in which the autonomous agent can be surrounded by up to six traffic vehicles, in which longitudinal and lateral safe distance must be maintained. Ideally, the agent would merge onto the highway, and overtaking occurs.
Is anyone willing to share an example or any tips that could give me a headstart?
Thanks!

请先登录，再回答此问题。

回答（1 个）

Salman Ahmed
Salman Ahmed 2021-11-17
Hi Christian,
I would recommend going to the following resources to help you start with:
  1. Driving Scenario Designer Basics : It allows you to Import ASAM OpenDRIVE® roads and lanes into a driving scenario and export road network, actors, and trajectories in a driving scenario to the ASAM OpenSCENARIO® 1.0 file format. In the ASAM specification you can find longitudinal and lateral distance actions.
  2. Example : Create Roads with Multiple Lane Specifications Using Driving Scenario Designer : This example shows how to interactively create roads with multiple lane specifications using the driving scenario designer app.
  3. Example : Highway Lane Change : This example shows how to simulate an automated lane change maneuver system for a highway driving scenario.
Hope it helps.

类别

在 Help Center 和 File Exchange 中查找有关 Automated Driving Toolbox 的更多信息

标签

产品


版本

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by