This module provides localization services. There are two ways in which localization is provided:
- The RTK (Real Time Kinematic) based method which incorporates GPS and IMU (Inertial Measurement Unit) information
- The multi-sensor fusion method which incorporates GPS, IMU, and LiDAR information.
modules/localization/
├── BUILD
├── common // common function, data structure and gflags definition
├── conf
├── cyberfile.xml
├── dag
├── launch
├── msf // msf localization implementation
├── ndt // ndt localization implementation
├── proto
├── README_cn.md
├── README.md
├── rtk // rtk localization implementation
└── testdata
- apollo::localization::MSFLocalizationComponent
- apollo::localization::NDTLocalizationComponent
- apollo::localization::RTKLocalizationComponent
In the provided RTK localization method, there are two inputs:
- GPS - The Global Positioning System
- IMU - Inertial Measurement Unit
In the multi-sensor fusion localization method, there are three inputs:
- GPS - The Global Positioning System
- IMU - Inertial Measurement Unit
- LiDAR - Light Detection And Ranging Sensor
An object instance defined by Protobuf message LocalizationEstimate
, which can be found in file localization/proto/localization.proto
.
cyber_launch start modules/localization/launch/msf_localization.launch
### Output
An object instance defined by Protobuf message `LocalizationEstimate`, which can be found in file `common_msgs/localization_msgs/localization.proto`.
### configs
| file path | type / struct | Description |
| ------------------------------------------------------------- | ------------------------------------------- | -------------------------- |
| `modules/localization/conf/localization_config.pb.txt` | `apollo::localization::LocalizationConfig` | localization type config |
| `modules/localization/conf/navi_localization_config.pb.txt` | `apollo::localization::LocalizationConfig` | localization type config |
| `modules/localization/conf/rtk_localization.pb.txt` | `apollo::localization::rtk_config::Config` | rtk mode config |
| `modules/localization/conf/localization.conf` | `gflags` | gflags config |
### Flags
| flagfile | type | Description |
| ---------------------------------------------------- | ---- | ------------------------------ |
| `modules/localization/common/localization_gflags.cc` | `cc` | localization flags header |
| `modules/localization/common/localization_gflags.h` | `h` | localization flags define |
### How to Launch
```bash
cyber_launch start modules/drivers/camera/launch/rtk_localization.launch // run rtk localization
cyber_launch start modules/drivers/camera/launch/ndt_localization.launch // run ndt localization
cyber_launch start modules/drivers/camera/launch/msf_localization.launch // run msf localization
Currently the RTK based localization is implemented in class RTKLocalization
. If you are implementing a new localization method, for example, FooLocalization
, you would need to perform the following steps:
-
In
localization_config.proto
, add a valueFOO
to theLocalizationType
enum type -
Go to the
modules/localization
directory, and create afoo
directory. In thefoo
directory, implement the classFooLocalization
by following the code in theRTKLocalization
class in thertk
directory.FooLocalization
has to be a subclass ofLocalizationBase
. Also create a filefoo/BUILD
by following the filertk/BUILD
-
You need to register the
FooLocalization
class in the functionLocalization::RegisterLocalizationMethods()
, which is located in thelocalization.cc
file. You can register it by inserting the following code at the end of the function:
localization_factory_.Register(LocalizationConfig::FOO, []()->LocalizationBase* { return new FooLocalization(); });
-
Make sure your code compiles by including the header files that defines class
FooLocalization
-
Now you can go back to the
apollo
root directory and build your code with commandbash apollo.sh build
in source env; whilebuildtool build -p modules/localization
in package management env.
For more information, refer to
Guowei Wan, Xiaolong Yang, Renlan Cai, Hao Li, Yao Zhou, Hao Wang, Shiyu Song. "Robust and Precise Vehicle Localization Based on Multi-Sensor Fusion in Diverse City Scenes," 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, QLD, 2018, pp. 4670-4677. doi: 10.1109/ICRA.2018.8461224. link