Braitenberg_vehicle_behavior

This project implements a Braitenberg vehicle simulation with both teleoperation and autonomous navigation capabilities. The robot uses sensors to detect obstacles in a maze environment and implements cross-wired Braitenberg behavior for obstacle avoidance. The simulation allows for parameter tuning to analyze the impact of sensitivity on navigation performance.

Features

• Manual teleoperation using arrow keys
• Autonomous navigation with Braitenberg vehicle behavior
• Sensitivity analysis with performance metrics
• Visualization of sensor readings and robot path
• Containerized application using Docker

Teleoperation Behavior

• Use arrow keys to manuever
  Up: ⬆️ Down: ⬇️Left: ⬅️Right: ➡️

Autonomous Behavior

The robot implements a cross-wired Braitenberg vehicle behavior where:

• Two sensors positioned at 45° angles detect obstacles
• Sensor readings influence turning behavior (right sensor affects left wheel, left sensor affects right wheel)
• The robot slows down when approaching obstacles
• The sensitivity parameter controls how strongly the robot responds to sensor readings

Sensitivity Parameter

The sensitivity parameter (0.2-1.0) determines how strongly the robot reacts to detected obstacles:

• Higher sensitivity: More aggressive turning and earlier response to obstacles
• Lower sensitivity: More gradual turning and later response to obstacles

The analysis feature tests different sensitivity values to find the optimal setting that minimizes both completion time and collision count.

Performance Metrics

The sensitivity analysis evaluates robot performance using:

• Completion time: How long it takes to navigate from start to finish
• Collision count: Number of times the robot collides with maze walls
• Performance score: A weighted combination (time + collisions*2) where lower is better

The analysis generates graphs showing how these metrics vary with different sensitivity values.

Troubleshooting

• If you encounter display issues, ensure X11 forwarding is properly configured
• If the container exits immediately, check that pygame is properly installed
• For performance issues, try adjusting the robot speed and sensor range parameters

How to run this repository

Install docker

sudo apt update
sudo apt install docker.io
sudo usermod -aG docker $USER   # Add yourself to the docker group
newgrp docker  # Apply group change (or logout & re-login)

Clone this repository

git clone git@github.com:keyurborad5/Braitenberg_vehicle_behavior.git

Build the docker Image

cd ~/Braitenberg_vehicle_behavior/
#Look for atleast these three files in the folder
#my_teleoperation.py, my_autonomous.py and Dockerfile
ls 
docker build -t simulation_image:latest .
# Check for the built image named: simulation_image:latest
docker images
#now you allow docker to connect with your X server
xhost +local:docker

Run the container with the application

# First we will run Teleoperation in our container
docker run -it --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --name teleop_container simulation_image:latest my_teleoperation.py
# Set the robot speed or just press Enter for DEFAULT speed
# Set the Robot angular speed or just Press Enter for DEFAULT speed
# Use Arrow keys to Teleoperate the robot and you can see the logs publishing on the terminal as well
# Use the following keys to control the robot:
# Up: ⬆️
# Down: ⬇️
# Left: ⬅️
# Right: ➡️

# Once done exploring the Teleoperation just CLOSE the GUI
# Now, We will run Autonomous behaviour in our contrainer
docker run -it --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --name autonomous_container simulation_image:latest my_autonomous.py
# Now terminal will ask some user input values. 
# Enter it of your choice or press ENTER for DEFAULT values selection
# Enter Robot Max Speed (default 5): 5
# Enter Turning Rate in degrees/unit time (default 2): 1
# Enter Sensor Range (default 150): 200
# Enter Sensitivity Factor (0.2-1.0, default 0.6): 0.8

Once Start the ROBOT will start to move autonomously based on the above set parameter.

EXIT

Container will be closed once the robot reaches the goal location or you can close it by closing the GUI application

Starting Analysis

Here I have tried to automate the analysis of the robot in a given maze to reach at the goal location in given predefined time that is 90 sec. Parameter I am varying for this analysis is sensitivity, which is being varied from 0.2 to 1.

# Start the Autonomous behavious once again
docker run -it --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --name autonomous_container simulation_image:latest my_autonomous.py

# Now press A from your keyboard

It will start the analysis and iterate it for 5 different sensitivity values with given set speed and angular speed.

EXIT

Once all the 5 iterations are complete a report will be printed on the terminal as well as a .png file will also be generated inside the container showing the graphical interpretation

NOTE : DONOT CLOSE THE GUI BEFORE COPYING THE GRAPH IMAGE

# To downlaod this analysis graph image to your local system open another terminal and go to your desired folder
docker cp autonomous_container:/app/sensitivity_analysis.png ./sensitivity_analysis.png

# you will get this output depending on your folder location.
Successfully copied 84kB to /home/ubuntu/sensitivity_analysis.png

Now you can access this image from your local system. Now exit by closing the GUI.