Kinova robotic arm
Kinect Arzue depth camera
Please click here to view the driver files of the following robots, as well as the driver files of the computer and communication board
Mechanical arm drive
Kinect Azure Camera driver
Pan and tilt drive
Site start-up
all_test.sh
The robot can automatically navigate to the desktop to clean up garbage in the home environment:
YOLOV5 identify the type of object, grab it and throw it to the trash can or put it back
Use the point cloud, calculate the depth information of bounding_box pixels to process the average, and get the position coordinates of the center of gravity point of the object relative to the camera, publish its tf
The basic functions of navigation goto (goal) and speech are defined
The robot human is defined so that navigation, robotic arm visual grasping and speech functions can be integrated
gpd_ros package
object_position.py
object_tf.cpp
Yolov5_ros yolov5.launch
res.launch
slam.launch
navigation.launch
GPSR: Difficult
true_follwer.py
GPES_find objects.py
take the object to placement2
find the object which we need
person how many |gender|
what is that.sh
The second line needs to start conda pytorch separately
The right hand is placed on the left shoulder to activate the follower
Raising your arm upwards activates the recognition function, which then points to the object you want it to broadcast.
The robot calculates the point of your finger, follows you and broadcasts the type of object you are pointing to and features such as color
Open door.sh
Open door.py
Receptionist.sh
Receptionist.py
The task flow of entertaining guests:
-
The robot arm opens the door,welcome the guests,ask for the guest's name and preferred drink
-
The robot will automatically navigate to the designated location
-
introduce the guest's name and favorite drink to the host, and guide the guest to an empty seat
mian class init
navigation--self.goto(0,0,0)
Azure subscriptions expire
The rviz node of the camera --notic same node name
Multithreading calls do not interrupt _thread.start_new_thread(text_to_speech, ("task begin!" ))
You can also just call text_to_speech("hurry").
Text-to-speech speech_to_text returns text
TF tree view error --rosrun rqt_tf_tree rqt_tf_tree
Node versus topic check --rosrun rqt_graph rqt_graph
Save the map:
rosrun map_server map_saver -f my2024_map
python3 '/home/xxx/scripts/teleop_twist_keyboard/teleop_twist_keyboard.py'
The whole point of github is to be able to look at commits and see what's different
用github的目的是可以查看提交的修改记录，找到每次不同 感谢2024机器人中心所有成员的努力 感谢往届学长的至关重要的帮助