Merge pull request #127 from AutoMecUA/dev

New release

.gitignore

@@ -33,7 +33,7 @@ MANIFEST
 *.manifest
 *.spec
 
-# Installer logs
+# Installer log
 pip-log.txt
 pip-delete-this-directory.txt
 

README.md

@@ -1,250 +1,10 @@
 # AutoMec-AD
 
-This project goal is to develop a fully autonomous car to compete in the National Robotics Festival of Portugal. This repository is where the main code of University of Aveiro car is hosted. 
+This repository serves as the main repository for the autonomous RC car team of Automec. The repository consist of ML code
+for the lane detection, template matching code for signal recognition, all incorporated in a ROS framework.
 
-![alt text](https://i.imgur.com/7FCETQ1.png)
-
-# Dependencies
-Make sure you have these installed before booting 
-
-  - Python3
-  - Ubuntu 20.04
-  - Ros Noetic
-  - numpy==1.20.1
-  - opencv-python==4.5.1.48
-  - matplotlib==3.3.4
-
-# How to run
-The code used in this repository is meant to drive a real world car you can do so by folliwing the instructions on [Manual Driving](https://github.com/DanielCoelho112/AutoMec-AD/tree/readme/core/src/ManualDriving#manual-driving), however using ROS we can simulate an enviroment where we can test our robot (car). To do so, continue following the Setting up ROS Enviroment section.
-
-First we need to create a catkin **workspace** to build our project. Choose an apropriate location for this.
-
-## Setting up ROS Enviroment
-```bash
-$ mkdir -p ~/catkin_ws/src
-$ cd ~/catkin_ws/
-$ catkin_make
-```
-Next you need to source catkin to your setup.bash
-
-If you are using bash terminal
-
-```bash
-$ source ~/catkin_ws/devel/setup.bash
-```
-
-Using zsh
-```bash
-$ source ~/catkin_ws/devel/setup.zsh
-```
-
-Next move in to the catkin src folder if you havent already and clone the repo.
-```bash
-cd catkin_ws
-cd src
-git clone https://github.com/DanielCoelho112/AutoMec-AD.git
-git checkout dev
-cd ..
-
-```
-Now you have the **main code** of our application but we still need to add some extra dependencies, ​this new repository that includes the turtle bot file and the arena used to simulate.
-
-```bash
-cd catkin_ws
-cd src
-git clone https://github.com/ROBOTIS-GIT/turtlebot3_msgs
-git clone https://github.com/ROBOTIS-GIT/turtlebot3_simulations
-git clone https://github.com/ROBOTIS-GIT/turtlebot3
-git clone https://github.com/callmesora/AutoMec-Deppendencies
-git clone https://github.com/prolo09/pari_trabalho3
-cd ..
-
-```
-
-All that's left to do is to run catkin_make, bhis will build your ros packages and install any depencies they have. When running catkin_make any dependencies installed in previous packages will still run in your new ones so be carefull when creating a new package
-.
-
-```bash
-catkin_make
-```
-
-Add the following to your .bashrc  file
-```bash
-export TURTLEBOT3_MODEL=waffle_pi 
-```
-
-
-If you don't know how to edit your .bashrc file type
-
-```bash
-nano ~/.bashrc
-```
-
-And add the previous line to the end of your file. Same procedure if you use zsh terminal but with .zshrc
-To Save press Cntrl+O , Enter . Cntrl+X to exit
-
-# Ackerman dependencies
-    sudo apt-get install ros-noetic-ros-controllers ros-noetic-ackermann-msgs
-
-# Launch with ackerman
-    roslaunch ackermann_vehicle_gazebo ackermann_robot_with_arena_conversion.launch
-
-# Running the simulation enviroment
-
-Execute the following on one terminal 
-```bash
-roslaunch robot_bringup bringup_gazebo.launch
-roslaunch robot_bringup spawn.launch
-```
-On a secound terminal.
-
-```bash
-
-roslaunch robot_bringup spawn.launch
-```
-First command will launch gazebo, secound one will spawn the robotcar. 
-After this you should see gazebo opening with the racing track as shown.
-![Gazebo](https://i.imgur.com/w7EFh7k.png)
-
-# How to drive 
-To test drive the car  type in gazebo  run
-```bash
-rqt
-```
-Now head down to pluggins --> Robot Tools --> Robot Stearing and select /robot/cmd_vel as it's topic.
-
-![Driving](https://i.imgur.com/ME4mgl7.png)
-
-After this we need to add the camera Go to  **Plugins --> Vizualization --> Image View** and select  */robot/camera/rgb/image_raw*
-
-# Vision Code
-During this project multiple aproaches are being tested , raw computer vision with opencv and a Machine Learning aproach. 
-
-## Lane_Recognition
-To test. cd into /core/src/VisionCode/Lane_Recognition
-Launch gazebo and the robot as in the **Running the simulation** section
-
-and run the script
-
-```bash
-rosrun core gazebo_lines.py
-```
-
-You should see multiple camera windows with different filters show up.
-
-## Signal Recognition
-To test. cd into /core/src/VisionCode/Signal Recognition
-
-Open a terminal and run
-
-```bash
-roscore
-```
-On a secound terminal
-```bash
-python3 SignalRecognition.py
-```
-
-
-# Code structure
-This project uses ROS as it's building blocks. It's divided into 4 packages
-
-## core 
-
-This package is divided into 4 folders.
-
-ArduinoCode: All arduino code made.
-
-VisionCode: All attempts at vision codes made.
-
-ManualDriving: Only contains the files for manual driving to be possible, latest version of the Arduino, twist converter.
-
-CommonFiles: Files that do not fit elsewhere, initialization files, etc.
-
-## robot_bringup
-Code to launch the gazebo and the robot in the simulation
-
-## robot_core
- ​
-Vision code ready to be implemented in gazebo
-
-Automatic driving using a ML Aproach
-
-## robot_description
-
-All files regarding robot description and stats
-
-
-
-
-
-# ML driving
-
-CATBOOST:
-          https://streamable.com/ol18mb
-
-          https://streamable.com/acich8
-
-
-          With images delivered to the catboost model:
-          https://streamable.com/kfi7j6
-
-CNN:
-          https://streamable.com/ysugtn
-
-## Dependencies for CNN
-    sudo apt install python3-pip
-
-    pip3 install opencv-python
-
-    pip3 install pandas
-
-    pip3 install sklearn
-
-    pip3 install tensorflow
-
-    pip3 install imgaug     
-
-
-## Get sample data for ml training
-
-roslaunch ackermann_vehicle_gazebo ackermann_robot_with_arena_conversion_mltrain.launch folder:=/cleantrack1
-
-É obrigatório o argumento folder:=/nome
-
-O folder tem de ser criado dentro da pasta "data"
-
-Lança tudo.
-
-- O mundo gazebo
-- O carro
-- O conversor twist para ackermann
-- O node de captura de dados
-- O node de rqt para conduzir o carro para as voltas de treino
-
-## Training ml model with sample data
-
-roslaunch robot_core training.launch folder:=/cleantrack1 model:=cleantrack1.h5
-
-São obrigatórios os argumentos:
-
-- folder:=/nome (não esquecer a barra no inicio)
-- model:=nome.h5 (Não esquecer a extensão .h5)
-
-## Driving with ml model
-
-Just drive, no signals.
-
-roslaunch ackermann_vehicle_gazebo ackermann_robot_with_arena_conversion_mlsolo.launch model:=cleantrack1.h5
-
-## Drive with two signals
-
-roslaunch ackermann_vehicle_gazebo ackermann_robot_with_arena_conversion_mlsignal.launch model:=cleantrack1.h5
-
-Não esquecer a extensão .h5
-
-
-
-## License
-https://streamable.com/t5thi0
+![alt text](https://raw.githubusercontent.com/AutomecUA/AutoMec-AD/main/images/car.jpeg)
 
+All the setup, commands and methods used are described in the [wiki](https://github.com/AutomecUA/AutoMec-AD/wiki). <br>
+This is still a WIP. Some errors are to be expected. If you have any doubts or want to report a bug,
+feel free to use the Issues or [send us an email](mailto:[email protected])!

cnn/scripts/cnn1/training1.py

@@ -110,6 +110,8 @@ def main():
 
     path = s + '/../../models/cnn1_' + modelname
 
+    enter_pressed = input("\n" + "Create a new model from scratch? [Y/N]: ")
+
     if enter_pressed.lower() == "y" or enter_pressed == "":
         model = createModel(image_width, image_height)
         is_newmodel = True

physical_bringup/launch/modules/signal_panel_recognition.launch

@@ -1,10 +1,12 @@
 <launch>
   <arg name="image_raw_topic" default="/real_camera"/>
   <arg name="signal_cmd_topic" default="/signal_vel"/>
+  <arg name="mask_mode" default="False"/>
 
   <include file="$(find signal_recognition)/launch/signal_panel_recognition.launch">
     <arg name="image_raw_topic" value="$(arg image_raw_topic)"/>
     <arg name="signal_cmd_topic" value="$(arg signal_cmd_topic)"/>
+    <arg name="mask_mode" value="$(arg mask_mode)"/>
   </include>
 
 </launch>

physical_bringup/launch/signal_usbcam.launch

@@ -2,10 +2,12 @@
     <arg name="camera_topic" default="/real_camera"/>
     <arg name="int_camera_id" default="2"/>
     <arg name="signal_cmd_topic" default="/signal_vel"/>
+    <arg name="mask_mode" default="True"/>
 
     <include file="$(find physical_bringup)/launch/modules/signal_panel_recognition.launch">
         <arg name="signal_cmd_topic" value="$(arg signal_cmd_topic)"/>
         <arg name="image_raw_topic" value="$(arg camera_topic)"/>
+        <arg name="mask_mode" value="$(arg mask_mode)"/>
     </include>
 
     <include file="$(find physical_bringup)/launch/modules/lane_camera.launch">

physical_bringup/launch/traxxas_drive_signal1.launch

@@ -12,6 +12,7 @@
 
     <arg name="camera_topic" default="/real_camera"/>
     <arg name="int_camera_id" default="2"/>
+    <arg name="mask_mode" default="False"/>
 
     <group if="$(eval arg('model') != '')">
 
@@ -35,6 +36,7 @@
         <include file="$(find physical_bringup)/launch/modules/signal_panel_recognition.launch">
             <arg name="signal_cmd_topic" value="$(arg signal_cmd_topic)"/>
             <arg name="image_raw_topic" value="$(arg camera_topic)"/>
+            <arg name="mask_mode" value="$(arg mask_mode)"/>
         </include>
 
         <include file="$(find physical_bringup)/launch/modules/lane_camera.launch">

physical_bringup/launch/traxxas_drive_signal2a.launch

@@ -12,6 +12,7 @@
 
     <arg name="camera_topic" default="/real_camera"/>
     <arg name="int_camera_id" default="2"/>
+    <arg name="mask_mode" default="False"/>
 
     <group if="$(eval arg('model') != '')">
 
@@ -35,6 +36,7 @@
         <include file="$(find physical_bringup)/launch/modules/signal_panel_recognition.launch">
             <arg name="signal_cmd_topic" value="$(arg signal_cmd_topic)"/>
             <arg name="image_raw_topic" value="$(arg camera_topic)"/>
+            <arg name="mask_mode" value="$(arg mask_mode)"/>
         </include>
 
         <include file="$(find physical_bringup)/launch/modules/lane_camera.launch">

requirements.txt

@@ -1,13 +1,8 @@
-controller_manager_msgs==0.19.4
-roslib==1.15.7
-sensor_msgs==1.13.1
-tensorflow==2.6.0
-cv_bridge==1.15.0
-rospy==1.15.11
+tensorflow==2.7.0
 pandas==0.25.3
-numpy==1.17.4
-tf==1.13.2
+numpy==1.19.2
 imgaug==0.4.0
 opencv_python==4.5.1.48
 matplotlib==3.1.2
 scikit_learn==0.24.2
+pillow==8.3.2

robot_driving/launch/physical_robot_joy.launch

@@ -0,0 +1,19 @@
+<launch>
+  <arg name="twist_dir_topic" default="/android_input_dir"/>
+  <arg name="bool_btn_topic" default="/android_input_velin"/>
+  <arg name="joy_topic" default="/joy"/>
+  <arg name="joy_number" default="0"/>
+  <arg name="deadzone" default="0"/>
+
+  <param name="joy_node/dev" value="/dev/input/js$(arg joy_number)"/>
+  <node name="joy" pkg="joy" type="joy_node">
+    <param name="deadzone" value="$(arg deadzone)"/>
+  </node>
+
+  <node name="robot_joy" pkg="robot_driving" type="joy_teleop.py" output="screen">
+    <param name="twist_dir_topic" value="$(arg twist_dir_topic)"/>
+    <param name="bool_btn_topic" value="$(arg bool_btn_topic)"/>
+    <param name="joy_topic" value="$(arg joy_topic)"/>
+  </node>
+
+</launch>