Welcome to ROSMASTER R2 repository

7. Voice control multi-point navigation

7.1. Function description

By interacting with the voice recognition module on ROSMASTER, you can use your voice to let ROSMASTER navigate to point 1/2/3 on an established map. The R2 button on the handle can cancel/enable this function at any time. .

7.2. Start

xxxxxxxxxx
#Raspberry Pi 5 master needs to enter docker first, please perform this step
#If running the script into docker fails, please refer to ROS/07, Docker tutorial
~/run_docker.sh

7.2.1. Function package path

xxxxxxxxxx
~/yahboomcar_ws/src/yahboomcar_voice_ctrl/

7.2.2. Calibrate the target point for voice navigation

robot side

xxxxxxxxxx
roslaunch yahboomcar_nav laser_bringup.launch #laser + yahboomcar

<PI5 needs to open another terminal to enter the same docker container

xxxxxxxxxx
roslaunch yahboomcar_nav yahboomcar_navigation.launch use_rviz:=false map:=house #Open navigation, change house to the map name for mapping

[use_rviz] Parameter: whether to open rviz.

[map] Parameters: map name, map to be loaded.

Turn on the visual interface (virtual machine side)

xxxxxxxxxx
roslaunch yahboomcar_nav view_navigate.launch

1. In the map rviz, adjust the initial posture of ROSMASTER;

2), terminal input

xxxxxxxxxx
rostopic echo /move_base_simple/goal

3. Use the 2D Nav Goal tool in rviz to give the car a target point in rviz, recorded as position 1. At this time, the coordinates of the target point will be printed out in the terminal window where you just viewed the data of /move_base_simple/goal. ,As shown below,

Just record the pose part of the data. Later, you need to send this part manually through the program.

4. Open ~/yahboomcar_ws/src/yahboomcar_voice_ctrl/scripts/voice_Ctrl_send_mark.py, and modify the pose data just recorded to the corresponding location.

xxxxxxxxxx
  pose.pose.position.x = 2.15381097794
  pose.pose.position.y = -5.02386903763
  pose.pose.orientation.z = 0.726492681307
  pose.pose.orientation.w = 0.687174202082

In addition to the two BC points, the position is also calibrated first, and then the calibrated value is modified into the function as an actual parameter.

7.2.2. Voice navigation

xxxxxxxxxx
roslaunch yahboomcar_nav laser_bringup.launch #laser + yahboomcar
roslaunch yahboomcar_nav yahboomcar_navigation.launch use_rviz:=false map:=house #Open navigation, change house to the map name for mapping
python ~/yahboomcar_ws/src/yahboomcar_voice_ctrl/scripts/voice_Ctrl_send_mark.py

After calibrating the initial pose in rviz, say "Hello, Xiaoya" to ROSMASTER to wake up the voice module. After hearing the voice module's feedback broadcast of "is", say to ROSMASTER "Navigate to position 1"; the voice module There will be a feedback message saying "Okay, heading to location 1". The same goes for other locations, as long as the coordinate values of the destinations of other points are written in the program.

7.3. Core code analysis

7.3.1. For the principle of navigation, please refer to the tutorial " 11. Lidar \ 12. Navigation and Obstacle Avoidance " The content of this section is mainly about judging the speech recognition results and packaging and sending

target point data,

xxxxxxxxxx
#Pack target point data
         speech_r = spe.speech_read()
         if speech_r == 19 :
             print("goal to one")
             spe.void_write(speech_r)
             pose.pose.position.x = 2.15381097794
             pose.pose.position.y = -5.02386903763
             pose.pose.orientation.z = 0.726492681307
             pose.pose.orientation.w = 0.687174202082
             pub_goal.publish(pose)
         elif speech_r == 20 :
             print("goal to show")
             spe.void_write(speech_r)
             pose.pose.position.x = 1.57744419575
             pose.pose.position.y = 4.8174996376
             pose.pose.orientation.z = -0.683335654604
             pose.pose.orientation.w = 0.730104364558
             pub_goal.publish(pose)
            
         elif speech_r == 21 :
             print("goal to three")
             spe.void_write(speech_r)
             pose.pose.position.x = -1.08106160164
             pose.pose.position.y = 1.30198049545
             pose.pose.orientation.z = -0.0132771070267
             pose.pose.orientation.w = 0.99991185533
             pub_goal.publish(pose)
            
         elif speech_r == 32 :
             print("goal to four")
             spe.void_write(speech_r)
             pose.pose.position.x = -1.08106160164
             pose.pose.position.y = 1.30198049545
             pose.pose.orientation.z = -0.0132771070267
             pose.pose.orientation.w = 0.99991185533
             pub_goal.publish(pose)

         elif speech_r == 33 :
             print("goal to Origin")
             spe.void_write(speech_r)
             pose.pose.position.x = -1.08106160164
             pose.pose.position.y = 1.30198049545
             pose.pose.orientation.z = -0.0132771070267
             pose.pose.orientation.w = 0.99991185533
             pub_goal.publish(pose)
         elif speech_r == 0 :
             pub_cmdVel.publish(Twist())S

7.3.2. Program flow chart

For complete code, please refer to:

xxxxxxxxxx
~/yahboomcar_ws/src/yahboomcar_voice_ctrl/scripts/voice_Ctrl_send_mark.py

7.3.3. Function module communication table