Scribd
Upload
16 views

Obstacle Recognition and Avoidance During Robot Navigation in Unknown Environment

This document presents two algorithms for obstacle recognition and avoidance during robot navigation in unknown environments. The first algorithm uses a Simulink model for navigation but cannot recognize repeated obstacles. The second algorithm proposes using standard deviations of laser scan vectors to recognize obstacles and break repetitive path loops. Experimental tests were conducted on a simulated Turtlebot robot in the Gazebo simulator. The algorithms aim to enable autonomous robot navigation with obstacle avoidance and recognition capabilities.

Uploaded by

Neerendra kumar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
16 views

Obstacle Recognition and Avoidance During Robot Navigation in Unknown Environment

This document presents two algorithms for obstacle recognition and avoidance during robot navigation in unknown environments. The first algorithm uses a Simulink model for navigation but cannot recognize repeated obstacles. The second algorithm proposes using standard deviations of laser scan vectors to recognize obstacles and break repetitive path loops. Experimental tests were conducted on a simulated Turtlebot robot in the Gazebo simulator. The algorithms aim to enable autonomous robot navigation with obstacle avoidance and recognition capabilities.

Uploaded by

Neerendra kumar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

International Journal of Engineering & Technology, 7 (3) (2018) 1400-1404

International Journal of Engineering & Technology


Website: www.sciencepubco.com/index.php/IJET
doi: 10.14419/ijet.v7i3.13926
Research paper

Obstacle recognition and avoidance during robot navigation in


unknown environment
Neerendra Kumar1* and Zoltán Vámossy2
1,2 John von Neumann Faculty of Informatics, Óbuda University, Budapest, Hungary.
1 Department of Computer Science & IT, Central University of Jammu, India.
* [email protected]

Abstract

In this paper, firstly, a model for robot navigation in unknown environment is presented as a Simulink model. This model is applicable
for obstacles avoidance during the robot navigation. However, the first model is unable to recognize the re-occurrences of the obstacles
during the navigation. Secondly, an advanced algorithm, based on the standard deviations of laser scan range vectors, is proposed
and implemented for robot navigation. The standard deviations of the lasers scans, robot positions and the time of scans with similar
standard deviations are used to obtain the obstacle recognition feature. In addition to the obstacle avoidance, the second algorithm
recognises the re-appearances of the obstacles in the navigation path. Further, the obstacle recognition feature is used to break the repeti-
tive path loop in the robot navigation. The experimental work is carried out on the simulated Turtlebot robot model using the Gazebo simulator.

Keywords: Gazebo simulator; Laser scan; Obstacle avoidance; Obstacle recognition; Robot navigation; Simulink.

1. Introduction igation model to navigating a robot in unknown environment is


presented in Section 2. Section 3 proposes an advanced solution for
Autonomous navigation of robots is gaining popularity among the obstacle avoidance using obstacle recognition. The Experimental
researchers day by day. Various requirements of obstacle avoidance set-up and the results are given in Section 4. Section 5 is for the
can be satisfied for the changing distance between robot and the conclusion of the work.
obstacles [1]. On the unmarked path, dynamic colour perception
model for the autonomous navigation is explained in [2]. Never- 2. Robot navigation model and problem defini-
theless, this model is not suitable for the laser range finder sensors.
tion
For a robot, motion planning with obstacle avoidance is given in
[3]. Using fusion of algorithms, an algorithm for navigation with
A robot navigation model with obstacle avoidance is given in Fig. 1.
obstacle avoidance is developed by [4]. Avoidance of unseen ob-
The robot and the navigation model communicate each other on robot
stacles studied in [5]. The complexity of the object matching task
operation system (ROS) topics using publisher-subscriber strategy.
rises for similar objects [6]. An algorithm for obstacles avoidance in
In Fig. 1, the Odom Subscriber block is subscribing the /odom topic
unknown environment using bumper events of the robot is proposed
of the ROS. Using the Odom Subscriber block the given navigation
in [7]. However, the algorithm presented in [7] has the limitation
model is receiving the positions of the robot during the navigation.
that the robot may struck in the corners and may follow the same
The ‘/scan’ Subscriber block of the navigation model is subscribing
loop of the path again and again. A fuzzy controller for obstacle
the /scan topic of the ROS. On the other hand, the robot is publishing
avoidance in unknown environment is presented in [8]. Although,
it’s pose on the ROS topic /odom and laser scan messages on the
the fuzzy rules, for the obstacles in the central part of the laser scan
ROS topic /scan. Therefore the robot positions and its lessor sensor
area, are right or left biased. Therefore, the robot may trap in a
readings are available to the robot navigation model presented by
repetitive navigation path loop in some situations. Using preceding
Fig. 1. The Publish geometry msgs/Twist block is publishing the
occurrences, user desired behaviour can be achieved during robot
linear and angular velocities, computed by the given model, on the
navigation [9]. Priority actions and early predictions based robot
ROS topic mobile base/commands/velocity. On the other side, the
navigation algorithm is presented in [10].
robot is subscribes the ROS topic mobile base/commands/velocity,
Although in the above studies, obstacle recognition based obstacle therefore, the robot is capable of receiving the linear and angular
avoidance is not exploited so far. Therefore, In this paper, obstacle velocities published by the given Simulink model. The Path positions
avoidance with obstacle recognition, to break the repetitive path loop data Writing block contains a MATLAB function to write the robot
during robot navigation in unknown and dynamic environment, is positions in a text file for the future use. The Algorithm 1 is used
presented. in MATLAB Function block of the Fig. 1. The input parameters of
Remaining part of this paper is structured as follows: A robot nav- the function defined in Algorithm 1 are corresponding to the input

Copyright © 2016 Author. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original work is properly cited.

You might also like