ros2+gazebo建立机器人

Building your own robot

ros2+gazebo建立机器人_第1张图片

In this tutorial we will learn how to build our own robot in SDFormat. We will build a simple two wheeled robot.本文用SDF文件建立一个2轮机器人

You can find the finished SDF file for the tutorial here.SDF文件点击下载

What is SDF

ros2+gazebo建立机器人_第2张图片

SDFormat (Simulation Description Format), sometimes abbreviated as SDF, is an XML format that describes objects and environments for robot simulators, visualization, and control.

SDF格式文件是一个用来描述仿真时候的各种配置的文件

Building a world

ros2+gazebo建立机器人_第3张图片

We will start by building a simple world and then build our robot in it. Open a new file called building_robot.sdf and copy the following code to it.



    
        
            0.001
            1.0
        
        
        
        
        
        
        

        
            true
            0 0 10 0 0 0
            0.8 0.8 0.8 1
            0.2 0.2 0.2 1
            
                1000
                0.9
                0.01
                0.001
            
            -0.5 0.1 -0.9
        

        
            true
            
                
                
                    
                    0 0 1
                    
                
                
                
                
                    
                    0 0 1
                    100 100
                    
                
                
                    0.8 0.8 0.8 1
                    0.8 0.8 0.8 1
                    0.8 0.8 0.8 1
                
                
            
        
    

Save the file, navigate to the directory where you saved the file and launch the simulator:

gz sim building_robot.sdf

Note: You can name your file any name and save it anywhere on your computer.

You should see an empty world with just a ground plane and a sun light. Check World demo to learn how to build your own world.

Building a model

ros2+gazebo建立机器人_第4张图片

Under the  tag we will add our robot model as follows:

Defining the model

ros2+gazebo建立机器人_第5张图片


    0 0 0 0 0 0

Here we define the name of our model vehicle_blue, which should be a unique name among its siblings (other tags or models on the same level). Each model may have one link designated as the canonical_link, the implicit frame of the model is attached to this link. If not defined, the first  will be chosen as the canonical link. The  tag is used to define the position and orientation of our model and the relative_to attribute is used to define the pose of the model relative to any other frame. If relative_to is not defined, the model's  will be relative to the world.

Let's make our pose relative to the world. The values inside the pose tag are as follows: X Y Z R P Y, where the X Y Z represent the position of the frame and R P Y represent the orientation in roll pitch yaw. We set them to zeros which makes the two frames (the model and the world) identical.

ros2+gazebo建立机器人_第6张图片

Every model is a group of links (can be just one link) connected together with joints.

Chassis

ros2+gazebo建立机器人_第7张图片

    
        0.5 0 0.4 0 0 0

We define the first link, the chassis of our car and it's pose relative to the model.

Inertial properties

ros2+gazebo建立机器人_第8张图片

     
        1.14395
        
            0.095329
            0
            0
            0.381317
            0
            0.476646
        
    

Here we define the inertial properties of the chassis like the  and the  matrix. The values of the inertia matrix for primitive shapes can be calculated using this tool.

Visual and collision

ros2+gazebo建立机器人_第9张图片

    
        
            
                2.0 1.0 0.5
            
        
        
        
            0.0 0.0 1.0 1
            0.0 0.0 1.0 1
            0.0 0.0 1.0 1
        
    

As the name suggests, the  tag is responsible for how our link will look. We define the shape of our link inside the  tag as a  (cuboid) and then specify the three dimensions (in meters) of this box inside the  tag. Then, inside the  tag we define the material of our link. Here we defined the  and  colors in a set of four numbers red/green/blue/alpha each in range [0, 1].

        
            
                
                    2.0 1.0 0.5
                
            
        
    

The  tag defines the collision properties of the link, how our link will react with other objects and the effect of the physics engine on it.

Note can be different from the visual properties, for example, simpler collision models are often used to reduce computation time.

After copying all the parts above into the world file in order, run the world again:

gz sim building_robot.sdf

Our model should look like this:

ros2+gazebo建立机器人_第10张图片

In the top left toolbar, click the Translate icon, then select your model. You should see three axes like this:

ros2+gazebo建立机器人_第11张图片

These are the axes of our model where red is the x-axis, green is the y-axis and blue is the z-axis.

Left wheel

ros2+gazebo建立机器人_第12张图片

Let's add wheels to our robot. The following code goes after the  tag and before the  tag. All the links and joints belonging to the same model should be defined before the .


    -0.5 0.6 0 -1.5707 0 0
    
        1
        
            0.043333
            0
            0
            0.043333
            0
            0.08
        
    

We defined the name of our link left_wheel and then defined its  relative_to the chassis link. The wheel needed to be placed on the left to the back of the chassis so that's why we chose the values for pose as -0.5 0.6 0. Also, our wheel is a cylinder, but on its side. That's why we defined the orientation value as -1.5707 0 0 which is a -90 degree rotation around the x-axis (the angles are in radians). Then we defined the inertial properties of the wheel, the mass and the inertia matrix.

Visualization and Collision

ros2+gazebo建立机器人_第13张图片

    
        
            
                0.4
                0.2
            
        
        
            1.0 0.0 0.0 1
            1.0 0.0 0.0 1
            1.0 0.0 0.0 1
        
    
    
        
            
                0.4
                0.2
            
        
    

The  and the  properties are similar to the previous link, except the shape of our link has the shape of  that requires two attributes: the  and the  of the cylinder. Save the file and run the world again, our model should look like this:

ros2+gazebo建立机器人_第14张图片

Right wheel

ros2+gazebo建立机器人_第15张图片



    -0.5 -0.6 0 -1.5707 0 0 
    
        1
        
            0.043333
            0
            0
            0.043333
            0
            0.08
        
    
    
        
            
                0.4
                0.2
            
        
        
            1.0 0.0 0.0 1
            1.0 0.0 0.0 1
            1.0 0.0 0.0 1
        
    
    
        
            
                0.4
                0.2
            
        
    

The right wheel is similar to the left wheel except for its position.

Defining an arbitrary frame

ros2+gazebo建立机器人_第16张图片

As of SDF 1.7 (Fortress uses SDF 1.8), we can define arbitrary frames. It takes two attributes:

  • name: the name of the frame
  • attached_to: the name of the frame or the link to which this frame is attached.

Let's add a frame for our caster wheel as follows:


    0.8 0 -0.2 0 0 0

We gave our frame name caster_frame and attached it to the chassis link, then the  tag to define the position and orientation of the frame. We didn't use the relative_to attribute so the pose is with respect to the frame named in the attached_to attribute, chassis in our case.

Caster wheel

ros2+gazebo建立机器人_第17张图片



    
    
        1
        
            0.016
            0
            0
            0.016
            0
            0.016
        
    
    
        
            
                0.2
            
        
        
            0.0 1 0.0 1
            0.0 1 0.0 1
            0.0 1 0.0 1
        
    
    
        
            
                0.2
            
        
    

Our last link is the caster and its pose is with respect to the frame caster_frame we defined above. As you could notice we closed the pose tag without defining the position or the orientation; in this case the pose of the link is the same as (identity) the frame in relative_to.

In the  and  tags we defined a different shape  which requires the  of the sphere.

ros2+gazebo建立机器人_第18张图片

We need to connect these links together; here comes the job of the  tag. The joint tag connects two links together and defines how they will move with respect to each other. Inside the  tag we need to define the two links to connect and their relations (way of movement).

Left wheel joint

ros2+gazebo建立机器人_第19张图片


    

Our first joint is the left_wheel_joint. It takes two attributes: the name name='left_wheel_joint' and the type type='revolute'. the revolute type gives 1 rotational degree of freedom with joint limits. The pose of the joint is the same as the child link frame, which is the left_wheel frame.

    chassis
    left_wheel

Every joint connects two links (bodies) together. Here we connect the chassis with the left_wheelchassis is the parent link and left_wheel is the child link.

    
        0 1 0 
        
            -1.79769e+308    
            1.79769e+308     
        
    

Here we define the axis of rotation. The axis of rotation can be any frame, not just the parent or the child link. We chose the y-axis with respect to the model frame so we put 1 in the y element and zeros in the others. For the revolute joint we need to define the  of our rotation angle in the  and  tags.

Note: The angles are in radians.

Right wheel joint

ros2+gazebo建立机器人_第20张图片

The right_wheel_joint is very similar except for the pose of the joint. This joint connects the right_wheel with the chassis.


    
    chassis
    right_wheel
    
        0 1 0
        
            -1.79769e+308    
            1.79769e+308     
        
    
Caster wheel joint

ros2+gazebo建立机器人_第21张图片

For the caster we need a different type of joint (connection). We used type='ball' which gives 3 rotational degrees of freedom.


    chassis
    caster

Conclusion

ros2+gazebo建立机器人_第22张图片

Run the world:

gz sim building_robot.sdf

It should look like this:

ros2+gazebo建立机器人_第23张图片

Hurray! We build our first robot. You can learn more details about SDFormat tags here. In the next tutorial we will learn how to move our robot around.

Video walk-through

ros2+gazebo建立机器人_第24张图片

A video walk-through of this tutorial is available from our YouTube channel: Gazebo tutorials: Building a robot.

你可能感兴趣的:(ros2)