Skip to content

jyotishp/multiagent-collision-avoidance

Repository files navigation

Multi-Agent Collision Avoidance

This implementation is based on the ideas from http://www.alonsomora.com/docs/18-alonsomora-TRO.pdf

Running the code

run.m is the main file that calls other functions in the directory. This file has the configurations related to the agents, axis limits for plots.

Note: If you are saving the plots to the disk as images, you need to create run directory before hand or the execution will fail. If there are images from a previous execution in this directory, those files will be overwritten.

Configuring the agents

The agents are declared at the beginning of run.m. For example, if you want to run the simulation with two agents, the array should more or less look like

agents = [
    addAgent('Agent-1', [-5 -5], [0 0], [ 5  5]),
    addAgent('Agent-2', [ 5  5], [0 0], [-5 -5])
]

Usage for addAgent function is

addAgent("Name of the agent", [initial position], [initial velocity], [goal position])

Configuring the plot

  • The axis limits of the plot are defined in run.m.
  • plotSimulation plots the current state of the agents along with their past path. This function is called in run.m#L51.

About

Decentralized multi-agent collision avoidance

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages