cbf-obstacle-avoidance

Control Barrier Function(CBF) implementation for quadrotor navigation

We use the standard CBF QP formulation,

$$ \begin{align*} \text{minimize} \quad & \frac{1}{2} |v - v_{\text{des}}|^2 \\ \text{subject to} \quad & \nabla h_i(x)^T v \geq -\alpha h_i(x) \quad \forall i \in \text{obstacles} \end{align*} $$

where, $h_i(x)$ is the distance to obstacle $i$

In this repository, this QP is solved using CVXOPT.

Results

Navigation in Pillar Worlds

Navigation in Sphere Worlds

Effect of varying the parameter $\alpha$

The parameter $\alpha$ in CBF significantly affects how aggressively the robot avoids obstacles

α = 0.1

α = 1.0

α = 10.0

Installation

Note that this code has only been tested with ROS-Noetic on Ubuntu 20.04. It is assumed that you already have a complete ROS installation. Also, ensure that CVXOPT is installed.

1. (optional) Set up a catkin workspace

   cd
   mkdir -p catkin_ws/src
   cd catkin_ws
   catkin init
   

2. Clone this repository in the workspace

   cd catkin_ws/src
   git clone https://github.com/Arcane-01/cbf-quadrotor-navigation.git .
   

3. Build the workspace

   cd ../
   catkin build
   source devel/setup.bash
   

Usage

1. To launch the Gazebo simulation with Bebop and CBF node, run the following command:

   bash run_cbf_quadrotor_navigation.sh <world_name> 
   

   • <world_name>: Gazebo world file from cbf/worlds (defaults to env4)

2. Once the CBF node has been initialized, you can begin navigation by setting a 2D Nav Goal in RViZ.

References

• Singletary et al., “Comparative Analysis of Control Barrier Functions and Artificial Potential Fields for Obstacle Avoidance” arXiv