Home
About
Welcome to the official documentation of Python Library tactics2d!
tactics2d
is an open-source Python library that provides diverse and challenging traffic scenarios for the development and evaluation of reinforcement learning-based decision-making models in autonomous driving. tactics2d
stands out with the following key features:
- Compatibility
- 📦 Trajectory dataset -- Enables seamless importation of various real-world trajectory datasets, including Argoverse, Dragon Lake Parking (DLP), INTERACTION, LevelX Series (highD, inD, rounD, ExiD), NuPlan, and Waymo Open Motion Dataset (WOMD), encompassing both trajectory parsing and map information.
- 📄 Map format -- Enables parsing and conversion of commonly used open map formats like OpenDRIVE, Lanelet2-style OpenStreetMap (OSM), and SUMO roadnet.
- Customizability
- 🚗 Traffic participants -- Supports the creation of new traffic participant classes with customizable physical attributes, physics dynamics/kinematics models, and behavior models.
- 🚧 Road elements -- Support the definition of new road elements, with a focus on regulatory aspects.
- Diversity
- 🛣️ Traffic scenarios -- Features an extensive range of built-in Gym-style traffic scenarios, including highway, lane-merging, unsignalized/signalized intersection, roundabout, parking, and racing.
- 🚲 Traffic participants -- Features a variety of built-in traffic participants with realistic physics parameters, detailed further here.
- 📷 Sensors -- Provides bird-eye-view (BEV) semantic segmentation RGB image and single-line LiDAR point cloud for model input.
- Visualization -- Offers a user-friendly visualization tool for real-time rendering of traffic scenarios and participants, along with the capability to record and replay traffic scenarios.
- Reliability -- Over 85\% code is covered by unit tests and integration tests.
What can you do with tactics2d
?
[Description]
Features
Updated on April 1, 2024.
Corresponds to version 0.1.6.
Dataset Parser
Support parsing maps and trajectories from the following datasets:
- HighD
- InD
- RounD
- ExiD
- Argoverse
- Dragon Lake Parking (DLP)
- INTERACTION
- NuPlan
- WOMD
Map Parser
Support parsing maps in the following formats:
- OpenStreetMap (OSM)
- OpenStreetMap annotated in Lanelet2
- OpenDRIVE (XODR)
Math Interpolation Algorithms
Support the following interpolation algorithms:
- B-Spline
- Bezier
- Cubic
- Spiral
- Dubins
- Reeds Shepp
Traffic Participant
The following traffic participants are implemented:
- Vehicle
- Cyclist
- Pedestrian
For each traffic participants, a set of parameters are available to configure the behavior.
Physics Model
The following physics model of traffic participants are supported:
- Bicycle model (Kinematic): recommended for cyclists and low-speed vehicles
- Bicycle model (Dynamic): recommended for cyclists and high-speed vehicles
- Point mass (Kinematic): recommended for pedestrians
- Single-track drift model (Dynamic): recommended for vehicles
Road Element
The following road elements are implemented:
- Lane
- Area
- Junction
- Road line
- Base class of traffic regulations
Traffic Event Detection
- Static collision detection
- Dynamic collision detection
- Arrival event detection
Sensor
- Bird-eye-view (BEV) semantic segmentation RGB image
- Single-line LiDAR point cloud
Why tactics2d
?
Similar Works
tactics2d
is crafted to offer a robust and intuitive environment tailored for the development and evaluation of autonomous driving decision-making models. As a third-party library, tactics2d
does not cater to any specific dataset; instead, its focus lies in facilitating parsing, visualization, log replaying, and interactive simulation across a diverse array of datasets and map formats. The table below provides a comparative overview of tactics2d
alongside other open-source simulators under active maintenance.
These tables are updated on 2024-04-01. Notations: = Implemented and tested = Under development
-
= Not implemented and not planned
Functionality
Simulator | Built-in RL Environment | Custom Trajectory | Custom Map | Dataset Compatibility | Interactive NPCs | Multi-agent |
---|---|---|---|---|---|---|
SUMO | - | - | - | |||
CarRacing | - | - | - | - | - | |
CARLA | - | - | ||||
CommonRoad | - | |||||
highway-env | - | - | - | - | ||
SMARTS | - | - | ||||
MetaDrive | - | |||||
NuPlan | - | - | - | - | - | |
InterSim | - | - | - | |||
TBSim | - | - | - | |||
Waymax | - | - | - | |||
Tactics2D |
Dataset Compatibility
tactics2d
excels in parsing various datasets into unified map and traffic participant data structures, enabling seamless integration for both log replay and interactive simulations.
Below is a comparison of dataset support between tactics2d
and other simulators. tactics2d
strives to accommodate a wide range of datasets, and we commit to keeping the table updated on a regular basis.
TODO
We have a plan to add support to the following datasets in the future:
- NGSIM
- Lyft 5
Feel free to suggest additional trajectory datasets to be incorporated into our support list by either opening an issue or submitting a pull request. We value community input and are committed to expanding our dataset coverage to better serve our users.
Simulators | Argoverse | DLP | INTERACTION | LevelX | NuPlan | WOMD |
---|---|---|---|---|---|---|
SUMO | - | - | - | - | - | - |
CarRacing | - | - | - | - | - | - |
CARLA | - | - | - | - | - | - |
CommonRoad | - | - | - | - | ||
highway-env | - | - | - | - | - | - |
SMARTS | - | - | - | - | - | - |
MetaDrive | - | - | - | - | - | - |
NuPlan | - | - | - | - | - | |
InterSim | - | - | - | - | ||
TBSim | - | - | - | - | ||
Waymax | - | - | - | - | - | |
Tactics2D |
Map Format Compatibility
Simulators | OpenDRIVE | OpenStreetMap | SUMO Roadnet |
---|---|---|---|
SUMO | |||
CarRacing | - | - | - |
CARLA | |||
CommonRoad | - | ||
highway-env | - | - | - |
SMARTS | - | ||
MetaDrive | - | ||
NuPlan | - | - | - |
InterSim | - | - | - |
TBSim | - | - | - |
Waymax | - | - | - |
Tactics2D |