Welcome to ROSMASTER X3 Plus repository

1 Multi-machine handle control

1.1 Multi-machine configuration

When using multi-machine handle control, it is first necessary to ensure that the robot is under the same local area network and configured with the same [ROS_MASTER_URI]; for multiple robots to control motion, there can only be one host. The example in this section sets the virtual machine as the host, and other robots as the slaves. There are several slaves. Of course, you can also set one robot as the master and others as the slaves.

1.1.1 Multi-machine communication settings

View the IP of the virtual machine

ifconfig 

Next, you only need to modify the .bashrc file of the slave(Robot side). There are several slave configurations.

xxxxxxxxxx
sudo vim ~/.bashrc 

Find the following line

xxxxxxxxxx
export ROS_MASTER_URI=http://IP:11311 

The [IP] here is the IP of the(virtual machine side).

xxxxxxxxxx
export ROS_MASTER_URI=http://192.168.2.106:11311 

After setting the IP, it is best to refresh the environment variables.

xxxxxxxxxx
source ~/.bashrc 

1.1.2 Multi-machine time synchronization

Install

xxxxxxxxxx
sudo apt install ntp ntpdate 

time check command

xxxxxxxxxx
date -R 

When there is a network, the system will automatically synchronize the network system time without setting. If there is no network, the time needs to be set when the time between robots is different

• have network

Commands to automatically synchronize network time

xxxxxxxxxx
ntpd pool ntp.ubuntu.com 

• No network

The test case is to use two robots.

Commands to manually set the time

xxxxxxxxxx
sudo date -s "2020-01-1 01:01:01" 

1) server side(host side) configuration

Open the [ntp.conf] file

xxxxxxxxxx
sudo vim /etc/ntp.conf 

add at the end of the file

xxxxxxxxxx
restrict  192.168.2.0  mask  255.255.255.0  nomodify  notrap 
server  127.127.1.0  # local clock 
fudge  127.127.1.0  stratum  10 

The first line is to enable the machine on the 192.168.2.xxx network segment to synchronize time with the local machine(specifically see if your ip is 192.168.2.xxx, if not, change it to your actual format) The second and third lines are to synchronize the hardware time of the machine with the ntp service of the machine.

After the change, restart the ntp service

xxxxxxxxxx
sudo /etc/init.d/ntp restart 

The server-side settings are completed, and the client-side settings are started below.

2) client(slave) configuration

Open the [ntp.conf] file

xxxxxxxxxx
sudo vim /etc/ntp.conf 

Add [server]+[IP] at the end of the file

xxxxxxxxxx
server 192.168.2.106 

3) time synchronization

On the [server side], execute the following code to check whether the [server side] ntp service is running

xxxxxxxxxx
service ntp status 

If not, restart the ntp service.

Network update(required for all devices)

xxxxxxxxxx
sudo /etc/init.d/networking restart 

time synchronization

xxxxxxxxxx
sudo ntpdate 192.168.2.106 

If the client's ntp is also enabled, you need to close the ntp service first, and then perform time synchronization.

xxxxxxxxxx
sudo /etc/init.d/ntp stop 

1.2 use

Take the virtual machine as the master and two robots as slaves as an example.

1.2.1 Turn on the robot

virtual machine side

xxxxxxxxxx
roscore 

Start the command(robot1 side), for the convenience of operation, this section takes [mono + laser + yahboomcar] as an example.

xxxxxxxxxx
#You need to enter docker first, perform this step more
#If running the script to enter docker fails, please refer to 07.Docker-orin/05, Enter the robot's docker container
~/run_docker.shroslaunch  yahboomcar_multi  laser_bringup_multi.launch  ns := robot1             # laser + yahboomcar 
roslaunch  yahboomcar_multi  laser_usb_bringup_multi.launch  ns := robot1         # mono + laser + yahboomcar 
roslaunch  yahboomcar_multi  laser_astrapro_bringup_multi.launch  ns := robot1    # Astra + laser + yahboomcar 

Start command(robot2 side), for the convenience of operation, this section takes [mono + laser + yahboomcar] as an example.

xxxxxxxxxx
#You need to enter docker first, perform this step more
#If running the script to enter docker fails, please refer to 07.Docker-orin/05, Enter the robot's docker container
~/run_docker.sh
roslaunch  yahboomcar_multi  laser_bringup_multi.launch  ns := robot2             # laser + yahboomcar 
roslaunch  yahboomcar_multi  laser_usb_bringup_multi.launch  ns := robot2         # mono + laser + yahboomcar 
roslaunch  yahboomcar_multi  laser_astrapro_bringup_multi.launch  ns := robot2    # Astra + laser + yahboomcar 

More bots and so on.

1.2.2 open the handle control

Method 1: One handle controls multiple robots at the same time

Connect the controller receiver to the virtual machine USB port and start the command

xxxxxxxxxx
roslaunch yahboomcar_multi joy_multi.launch 

Note: The [launch] file defaults to three robots.

joy_multi.launch

xxxxxxxxxx
< launch > 
    < arg  name = "first_robot1"  default = "robot1" /> 
    < arg  name = "second_robot2"  default = "robot2" /> 
    < arg  name = "third_robot3"  default = "robot3" /> 
    < param  name = "use_sim_time"  value = "false" /> 
    < node  name = "joy_node"  pkg = "joy"  type = "joy_node"  output = "screen"  respawn = "false" /> 
    <!--  ############################# first_robot1 ############################# --> 
    < include  file = "$(find yahboomcar_multi)/launch/library/joy_base.launch" > 
        < arg  name = "ns"  default = "$(arg first_robot1)" /> 
    </ include > 
    <!--  ############################# second_robot2 ############################# --> 
    < include  file = "$(find yahboomcar_multi)/launch/library/joy_base.launch" > 
        < arg  name = "ns"  default = "$(arg second_robot2)" /> 
    </ include > 
    <!--  ############################# third_robot3 ############################# --> 
    < include  file = "$(find yahboomcar_multi)/launch/library/joy_base.launch" > 
        < arg  name = "ns"  default = "$(arg third_robot3)" /> 
    </ include > 
</ launch > 

If there are 2 robots, just comment out the part of the third robot, and it will not affect if you do not comment out; if there are 4 robots or more, you can add them according to the first 3 samples.

Method 2: One handle controls one robot alone

Connect the controller receiver to the corresponding robot, and start the command on the corresponding robot side(take robot1 as an example)

<Open another terminal and enter the same docker container

xxxxxxxxxx
roslaunch yahboomcar_multi joy_each.launch ns:=robot1 

1.3 handle control analysis

Node view

xxxxxxxxxx
rqt_graph 

1.3.1 Controlling multiple robots

It can be seen from the figure that when a handle controls multiple robots at the same time, only one [joy_node] node is needed to receive the handle signal, and different [yahboom_joy] nodes are set for different robots to control the corresponding robot.

1.3.2 Controlling a robot

It can be seen from the figure that when a handle controls a robot, each robot is individually set and controlled.