rl_sar/README_CN.md

106 lines
2.3 KiB
Markdown
Raw Normal View History

2024-03-14 11:36:49 +08:00
# rl_sar
2024-03-07 11:54:33 +08:00
[English document](README.md)
2024-03-22 00:16:13 +08:00
四足机器人强化学习算法的仿真验证与实物部署。"sar"代表"simulation and real"
2024-03-14 11:36:49 +08:00
2024-03-07 11:54:33 +08:00
## 准备
拉取代码(同步拉取子模块)
```bash
2024-03-14 11:36:49 +08:00
git clone --recursive https://github.com/fan-ziqi/rl_sar.git
2024-03-07 11:54:33 +08:00
```
如有更新:
```bash
git pull
git submodule update --remote --recursive
```
在任意位置下载并部署`libtorch`
```bash
cd /path/to/your/torchlib
wget https://download.pytorch.org/libtorch/cpu/libtorch-cxx11-abi-shared-with-deps-2.0.1%2Bcpu.zip
unzip libtorch-cxx11-abi-shared-with-deps-2.0.1+cpu.zip -d ./
echo 'export Torch_DIR=/path/to/your/torchlib' >> ~/.bashrc
```
安装 `teleop-twist-keyboard`
```bash
sudo apt install ros-noetic-teleop-twist-keyboard
```
## 编译
自定义代码中的以下两个函数,以适配不同的模型:
```cpp
torch::Tensor forward() override;
torch::Tensor compute_observation() override;
```
然后到根目录编译
```bash
cd ..
catkin build
```
## 运行
2024-03-14 13:11:01 +08:00
运行前请将训练好的pt模型文件拷贝到`rl_sar/src/rl_sar/models`中
2024-03-14 11:36:49 +08:00
### 仿真
2024-03-07 11:54:33 +08:00
新建终端启动gazebo仿真环境
```bash
source devel/setup.bash
2024-03-29 16:49:28 +08:00
roslaunch rl_sar start_a1.launch
2024-03-07 11:54:33 +08:00
```
新建终端,启动控制程序
```bash
source devel/setup.bash
2024-03-14 13:13:47 +08:00
rosrun rl_sar rl_sim
2024-03-07 11:54:33 +08:00
```
新建终端,键盘控制程序
```bash
rosrun teleop_twist_keyboard teleop_twist_keyboard.py
```
2024-03-14 11:36:49 +08:00
### 实物
2024-03-07 11:54:33 +08:00
2024-03-18 14:53:31 +08:00
与实物的连接可分为无线与有线形式
* 无线连接机器人发出的Unitree开头的WIFI **(注意:无线连接可能会出现丢包断联甚至失控,请注意安全)**
* 有线用网线连接计算机和机器人的任意网口配置计算机ip为192.168.123.162网关255.255.255.0
2024-03-14 11:36:49 +08:00
新建终端,启动控制程序
2024-03-07 11:54:33 +08:00
2024-03-14 11:36:49 +08:00
```bash
source devel/setup.bash
2024-03-14 13:13:47 +08:00
rosrun rl_sar rl_real
2024-03-14 11:36:49 +08:00
```
2024-03-14 00:18:32 +08:00
2024-03-18 14:53:31 +08:00
按下遥控器的**R2**键让机器人切换到默认站起姿态,按下**R1**键切换到RL控制模式任意状态按下**L2**切换到最初的趴下姿态。左摇杆上下控制x左右控制yaw右摇杆左右控制y。
2024-03-22 00:16:13 +08:00
## 引用
如果您使用此代码或其部分内容,请引用以下内容:
```
@software{fan-ziqi2024rl_sar,
author = {fan-ziqi},
title = {{rl_sar: Simulation Verification and Physical Deployment of the Quadruped Robot's Reinforcement Learning Algorithm.}},
url = {https://github.com/fan-ziqi/rl_sar},
year = {2024}
}
```