Giter Club home page Giter Club logo

kobot's Introduction

1.Overview

ROS package for Kobot -- Heteregoneous swarm robot platform.

Kobot package is actively used and tested on RPi 3B+ running Ubuntu Mate 18.04, ROS Melodic and RPi Zero W running Raspbian Buster and ROS Melodic.

Additionally, this repository contains dotfiles (.vimrc, .zshrc, .tmux.conf), shell scripts and plain hardware test scripts that are used by Kobot robots.

Currently, 3 different swarm robots are extended from the Kobot platform:

  • Kobot-W (Wheeled robot, equipped with IR floor sensors, 9-Axis IMU, Xbee for local communication) Design Link
  • Kobot-F (Flying robot using Ryze Tello as the base) Desing Link
  • Kobot-T (Tracked Robot using Zumo from Pololu as the base, equipped with a 2-D.o.F. compliant gripper) [Design Link will be available soon]

SD Card Images

Raspberry Pi 3B/3B+

SD Card Image of the Kobot-W running Ubuntu Mate 18.04 and ROS Melodic

Raspberry Pi Zero W

SD Card Image of the Kobot-F running Raspbian Buster and ROS Melodic Image of the Kobot-T running Raspbian Buster and ROS Melodic (In-progress)

Branches

Branches are used for developing extended robot specific features until they are ready to be merged with the main branch. Maintainers and branches are:

Hardware Architecture

Alt text

Software Architecture

Alt text

Dependencies

If you are not using the SD Card images and would like to use kobot_ros package for other purposes, there are quite a few required dependencies which are given below (these are not updated regularly):

  • cv_bridge (need to build as a seperate catkin workspace)
  • camera_info_manager (required by ROS nodes)
  • tf (required by ROS nodes)
  • Python 3.6 or later version (not the default interpreter for ROS Melodic)
  • opencv-contrib-python (need to build from source for RPi Zero and 3B+)
  • mpu9250_jmdev (handling i2c comm. with MPU9255 9-Axis IMU)
  • adafruit_neopixel (handling comm. w/ neopixel led ring by using one of the GPIO pins)
  • pyserial (handling the serial comm. w/ Xbee RF Module)
  • smbus2 (handling all the i2c communication)
  • numpy (handling linear algebra)
  • RPi.GPIO (handling basic GPIO operations such as digital write, read)

Installation

There are no installation instructions for the stand-alone kobot package. The recommended way is using one of the SD Card images.

2.Nodes

This repository contains two types of nodes:

  • High-level beahvioural nodes to control the swarm behaviour of the robots
    • flocking Paper Link i.e. Swarm members perform alignment, avoidance and cohesion using robot, obstacle detection, range of the objects, heading with resepect to an absolute reference and non-holonomic velocity control Video Link
    • alignment, i.e. modified flocking algorithm where swarm members align their headings w.r.t. a common orientation reference
    • wall_following i.e. modified flocking algorithm where swarm members follow walls instead of avoiding perpendicularly
    • dispersion_avoidance, i.e. modified flocking algorithm where Kobot-F robots disperse in z-axis (altitude), avoid obstacles Video Link
    • beeclust Paper Link i.e. Swarm members aggregate on a cue denoted by high IR reflectivity (other abstractions are possible), avoid obstacles by random walk and wait in the cue when collide with another robot inside cue. Duration is determined by the reflectivity
    • lba (landmark-based aggregation) Paper Link i.e. Extended version of beeclust where swarm members learn and store the paths from landmarks to the cue (landmarks are abstracted by ArUco markers Paper Link detection by OpenCV) Video Link
    • rl_lba (landmark-based aggregation using reinforcement learning), i.e. Extended version of lba exploring, exploiting an action space when a landmark in the environment is detected
  • Sub-system driver nodes abstracting the hardware using ROS messages and subscriber/publisher architecture.
    • common nodes for all extended robots
      • range_n_bearing (get angle and range of objects and decide whether robot or obstacle)
      • landmark_pose_estimator (estimate pose of the ArUco marker abstracting landmarks)
      • pose_controller (controller pose of the robot, i.e. x,y,theta)
      • heading (get heading of the robot w.r.t. a common reference)
      • local_communication (broadcast locally using XBee RF module)
      • battery (check battery cell voltages regularly)
    • kobot_w, kobot_t specific nodes
      • teleop_key (tele-operate robot by using keyboard)
      • cmd_vel2motors (convert ROS's cmd_vel msg to seperate velocity commands for motors)
      • differential_driver (set, get controller params., velocity of the motors)
      • odom (collect wheel odometry data from the motors, get robot pose)
      • mpu9255 (get orientation data from IMU)
      • floor_sensor (get IR reflection intensity from floor sensors)
    • kobot_f specific nodes
      • tello_driver (to operate Ryze Tello drone)

Messages

Custom ROS messages used by the Kobot sub-systems and behavioural nodes:

  • range_n_robot, i.e. reading of an individual sensor of the range and bearing board. Robot field is a boolean which is true for robots and false for all other objects. Range field is the range reading corresponding to the sensor.
  • range_n_bearing, i.e. array of all sensor of the range_n_bearing board.
  • virtual_sensor, i.e. array of sensor readings.
  • floor_sensor, i.e. array of individual IR floor sensor readings.
  • landmark_pose, landmarks are abstracted by using Aruco markers,

Parameter Files

Three types of parameter files in .yaml format are used to configure the operation of the nodes:

  • Behavioural parameters
    • flocking
    • alignment (modified flocking beta=0, u_max=0)
    • wall_following (modified flocking avoid_angle = 90)
    • lba (modified beeclust)
    • beeclust (modified lba d_landmark = 0)
    • rl_lba (modified lba)
  • Calibration files, these files are specific to robots and stored locally hence, they are ignored by git.
    • IMU (gyro bias, magnetometer hard-iron, soft-iron biases)
    • Camera (intrinsic parameters, calibrated with a Charuco Board)
    • Odometry (left, right wheel radii and axle distance)
  • Controller
    • Velocity Controller (PID gains, velocity limits, control loop frequency)
    • Pose Controller (PID gains, error-window, max. translational velocity)

Important Parameters

Flocking Parameter Explanation Default Value
beta Weight of proximal control to heading alignment 1.5
u_max Max. translational velocity 0.1 m/s
sigma_des_robot Equilibrium distance from robots 350 mm
sigma_des_obstacle Equilibrium distance from obstacles 1000 mm
avoid_angle Angle of avoidance 180 deg
Aggregation Parameter Explanation Default Value
w_max Max. wait duration inside the cue 120 s
u_max Max. translational velocity 0.1 m/s
d_landmark Max. detection distance for landmarks 500 mm
d_obs Max. detection distance for obstacles 100 mm
d_rob Max. detection distance for robots 250 mm

Launch Files

Launch files are used for running multiple nodes, loading parameters to the ROS's parameter server and handling individual namespaces.

  • Sub-system launch files
    • move
    • heading
    • landmark_pose_estimator
    • floor_sensing
  • Behaviour launch files
    • flocking
    • alignment
    • wall_following
    • beeclust
    • lba
    • rl_lba
  • Calibration launch files
    • calibrate_camera
    • calibrate_imu

4.Work in Progress

  • Kobot-T sub-system drivers
  • flocking behaviour for Kobot-F
  • Kobot-H and Kobot-L (holonomic wheeled and legged swarm robots) conceptual design

5.License

MIT License

Copyright (c) 2021 cemoke

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

6.Additional Links

Related Repositories

Alt text

  • kobotMotorController Link
  • kobotRangeNBearing Link
  • kobot_base Link

Related Webpages

  • ROMER -- Instuition where Kobot platform is deleoped Webpage
  • Kovan Research Lab -- Laboratory where Kobot platform is developed Webpage

kobot's People

Contributors

cemoke avatar

Stargazers

Eren Bilaloglu avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.