Customer center

We are a boutique essay service, not a mass production custom writing factory. Let us create a perfect paper for you today!

Example research essay topic: A New Mobile Robot Navigation Algorithm - 1,655 words

NOTE: Free essay sample provided on this page should be used for references or sample purposes only. The sample essay is available to anyone, so any direct quoting without mentioning the source will be considered plagiarism by schools, colleges and universities that use plagiarism detection software. To get a completely brand-new, plagiarism-free essay, please use our essay writing service.
One click instant price quote

A New Mobile Robot Navigation Algorithm By using Fuzzy Logic Control and Virtual Target Technique Dr. Omid Reza Esmaeili Motlagh Intelligent Systems and Robotics (ISRL) Lab Institute of Advanced Technology, University Putra Malaysia (UPM), Malaysia Abstract A new fuzzy logic control system is developed for reactive navigation of a behavior-based mobile robot. The robot perceives its environment through an array of eight sonar sensors and self positioning-localization sensors. While the fuzzy logic body of the algorithm performs the main tasks of obstacle avoidance, target seeking and speed control, an actual-virtual target switch strategy enables the robot to wall following behavior when needed.

This significantly results in resolving the problem of limit cycles in any type of multiple dead end in local navigation which is an advantage beyond pure fuzzy logic approach and common virtual target techniques. In this work, multiple traps may have any shape or arrangement from barriers forming simple corners and U-shape dead ends to loops, maze, snail shape, and other complicated shapes. Under control of the algorithm, the robot makes logical trajectories, avoids any obstacle, and adjusts its speed efficiently for better obstacle avoidance and according to power considerations and actual limits. Final trajectory results are demonstrated by simulation work in compression with results from other related methods to show the effectiveness of the proposed approach. 1. Introduction Autonomous mobile robots defined as robots capable of intelligent motion and action without requiring either a guide to follow or a tele-operator control, which involves the integration of many different bodies of knowledge.

This makes mobile robotics a challenge worthwhile. Mobile robot local path planning or also called reactive navigation in an unknown and changing environment with uncertainties is one of the most challenging problems in robotics. For real-time autonomous navigation, the robot should be capable of sensing its environment, interpreting the sensed information to obtain the knowledge of its position and the environment, planning a real-time route from an initial position to a target with obstacle avoidance, and controlling the robot direction and velocity to reach the target [ 1 ]. There are many approaches to real-time local navigation that deal with one or more related issues and try to tackle the problem in their own way.

Some of the online approaches to path planning include wall following method [ 2 ], artificial potential field methods[ 3, 4 ], virtual target approach [ 5, 6 ], land mark learning [ 7 ], edge detection, graph-based methods, vector field histogram methods, dynamic window approaches, neural network based methods, fuzzy logic methods [ 8, 9 ], and many others. Since the introduction of the fuzzy logic control (FLC) in 1965 by Lot Zadeh of the University of California at Berkeley [ 10 ], this approach to mobile robot navigation and obstacle avoidance has been investigated by several researchers. Fuzzy systems have the ability to treat uncertain and imprecise information using linguistic rules, thus, they offer possible implementation of human knowledge and experience and have an advantage in that they do not require a precise analytical model of the environment. 1. 1. The problem of multiple dead end Although fuzzy logic control (FLC) technique can simplify navigation problems but there are situations happen in local work space where a pure fuzzy logic approach fails in taking appropriate action. Of these troublesome situations obstacles forming a loop shape also called dead end traps are the most common. The local minima situation occurs when a robot navigating past obstacles towards a target with no prior knowledge of the environment gets trapped in a loop.

This happens if the environment consists of concave obstacles, and the like. Fig. 1 shows a robot with pure fuzzy logic navigator getting trapped in a U-shape dead end. Because rules that are fired for target attractor and obstacle repulsed modules give output actions that neutralize each other, the robot gets into an infinite loop or local minima [ 7 ]. Fig. 1. Inability of fuzzy logic control in Local Minimum Initially the robot moves directly toward the target due to target seeking behavior, because there is no obstacle sensed in front of the robot and this is an ideal shortest path to the target. This continues until point A Where the robot detects obstacles at the direct front, it makes a right turn due to obstacle avoidance behavior that results to wall following until robot reaches the point C.

This is because until this point both the target and obstacle are at the left hand side of the robot. But as the robot in passing by the point C the target is going to be at right hand side of the robot, while the obstacle is still at the left hand side. Therefore at the point D due to both target seeking and obstacle avoidance behaviors the robot goes back to ward the target. The result of this behavior is that the robot wanders indefinitely in the dead end trap called the limit cycle problem or local minimum. The situation even worsens when mobile robot encounters two or more dead ends in row forming a multiple minimum. In the next part few related works on minimum avoidance will be review and their draw backs in multiple minimum situations will be discussed.

Later in part 3 a new target switch approach to resolve the stated problem will be introduced. 2. Related work Other than pure fuzzy logic approaches there are many types of combinational navigation approaches, which range broadly according to two aspects of the extent to which their fuzzy logic navigator makes ultimate navigation decisions, and their degrees of dependence on memory data and artificial intelligence. 2. 1. Zhu and Yang state memory method [ 8 ] In this method the dead cycle problem is resolved by using a state memory strategy. The variables from which this method makes ultimate decision of either to keep turning around the obstacle (states 1 or 2) or to reach the target (state 0) are the robot current distance to the target (Dc) and the robot initial distance to the target (Dm) memorized when the obstacle was first detected (first change from state 0 to state 1 or 2). But this distance-based decision making results in poor trajectories in situations like here from point D toward the target. Since Dm was memorized when the robot was traveling from A to B, and the condition for changing the state to 0 or target seeking is; Dc shorter than Dm, therefore from point D the robot can not go straight toward the target and has to keep turning around the obstacle until point E where Dc becomes shorter than Dm.

Other than this, large size of the whole algorithm including a set of 48 rule bases and the assistant state memorizing algorithm make the method less interesting. Fig. 2. Yang method using distance-based memory states 2. 2. Krishna and kara landmark learning [ 7 ] This method is a real-time collision avoidance algorithm with the local minima problem resolved by classifying the environment based on the spatio-temporal sensory data sequences.

Although this method has a good result as shown in Fig. 3. a but it highly depends on the landmark recognition and therefore needs exact coordination localization. In addition, it is difficult to choose a correct direction to follow the wall boundary as seen in Fig. 3. b. [ 9 ]. Fig 3. Krishna and kara landmark learning approach 2. 3.

Wang and LIU minimum risk method [ 9 ] This method is based on trial-and-return behavior phenomenon as shown in Fig. 4. a and 4. b. If there is no obstacle blocking the nearest exit, the minimum risk approach is guaranteed to find a nearest exit to escape from the local minimum and ultimately reach the goal. Therefore when the nearest exit in Fig. 4. a is blocked by a long wall based on this approach the nearest exit will be the opening at the right hand side where the wall ends.

And based on the same principle the robot can find its way out of the U-shape dead end in Fig. 4. b. Fig. 4. Wang minimum risk approach based on FL Although the minimum risk approach seems to be robust in traps of this type but obviously there is a major problem with trial and return-based motion because for a mobile robot Power consumption And spent time from start to target are the two important issues that are totally ignored in this approach.

This is shown in Fig. 5 with an example of a mobile robot trying hardly to find its way out of a simple dead end. Fig. 5. trial and error navigation causes loss of time and power. 2. 4. Xu and Tso virtual target method [ 5 ] Has good properties in minimum avoidance behavior and the robot makes logical trajectories. (Fig. 6) Fig. 6. Xu. and Tso target switch scheme; virtually shifting the target &# 61552; &# 61472; radian to the opposite direction In this method the target is virtually relocated to points directed away from the true target, therefore after passing through the critical points C or F, the robot maintains the same tendency of turning to the left from C &# 61664; I &# 61664; J &# 61664; N or to the right from F &# 61664; L &# 61664; M &# 61664; N.

The virtual target orientation is given by tr = - (&# 61552; - tr 0) following the point C or tr = &# 61552; - tr 0 following the point F, where tr 0 and tr represent the real and virtual target orientations respectively, and the condition for the target switching is given by |tr 1 tr 0 | > TR, where tr 0 and tr 1 are the real target orientations at the two consecutive reaction instants and TR is a threshold for the abrupt change in the target orientation. Due to the actual t...


Free research essays on topics related to: dead end, obstacle, target, left hand side, robot

Research essay sample on A New Mobile Robot Navigation Algorithm

Writing service prices per page

  • $18.85 - in 14 days
  • $19.95 - in 3 days
  • $23.95 - within 48 hours
  • $26.95 - within 24 hours
  • $29.95 - within 12 hours
  • $34.95 - within 6 hours
  • $39.95 - within 3 hours
  • Calculate total price

Our guarantee

  • 100% money back guarantee
  • plagiarism-free authentic works
  • completely confidential service
  • timely revisions until completely satisfied
  • 24/7 customer support
  • payments protected by PayPal

Secure payment

With EssayChief you get

  • Strict plagiarism detection regulations
  • 300+ words per page
  • Times New Roman font 12 pts, double-spaced
  • FREE abstract, outline, bibliography
  • Money back guarantee for missed deadline
  • Round-the-clock customer support
  • Complete anonymity of all our clients
  • Custom essays
  • Writing service

EssayChief can handle your

  • essays, term papers
  • book and movie reports
  • Power Point presentations
  • annotated bibliographies
  • theses, dissertations
  • exam preparations
  • editing and proofreading of your texts
  • academic ghostwriting of any kind

Free essay samples

Browse essays by topic:

Stay with EssayChief! We offer 10% discount to all our return customers. Once you place your order you will receive an email with the password. You can use this password for unlimited period and you can share it with your friends!

Academic ghostwriting

About us

© 2002-2024 EssayChief.com