INTRODUCTION
Chapter 2 LITERATURE REVIEW
2.2 Different Approaches used for Path Planning and Navigation Control of Mobile Robots
2.2.6 Sensor based Navigation Approaches
For years, many researchers have analyzed on control techniques of mobile robots equipped with various types of sensors.
2.2.6.1 Navigation using Sensor based Approaches
Zelinsky [81] has discussed about the mobile robot map making using sonar sensors. His robot is capable of producing high resolution maps of an outdoor environments equipped with sonar range finding sensors. His robot can recognize the obstacle distances using sonar sensors. He has given various exercises of his mobile robot equipped with sonar sensors.
Varadarajan et al. [82] have developed a technique to fuse the sonar and wheel encoder information to produce a map of an environment. In their methodology, they have used specular environment which causes the sonar to produce completely unusable readings which often occurs with smooth surfaces. Masek et al. [83] have discussed about obstacle sensing using mobile robot equipped with a ring of sonar. In their method of sonar ring design, multi-wave superposition has been taken care to attain a quasi- homogeneous sonar beam intensity pattern. They have claimed that using their sonar sensor the robot can accurately detect the obstacles during navigation.
Maris [84] has discussed about attention based mobile robot using a reconfigurable sensor. The mobile robot designed by Harper and Phillip [85] can navigate through many environments that contain plants. They have used a sensor that can recognize plants which will be very useful for navigation in those environments. They have used an ultrasonic sensor to distinguish various plants according to the structure of the plants.
Asharif et al. [86] have discussed about sensory data integration to obtain the grid map required for robot navigation. They have used neural network along with sensory data for path planning of the mobile robot.
Benet et al. [87] have used infrared (IR) sensors in their robots for obstacle avoidance. They have claimed that IR sensor can accurately measure distances with reduced response time. They have given experimental results to validate the theory developed. A sensing system using sonar ring has been used by Ming et al. [88] for obstacle detection and avoidance. To eliminate the error from the sonar ring they have used a neural filtering method to prevent cross talks between the sensors.
Carelli et al. [89] have discussed about control, navigation and wall following behavior of a mobile robot using sonar and odometric sensorial information. They have added the OA capability to the control system as a perturbation signal. The paper by Carmena and John [90] discusses about biologically inspired engineering with the use of narrowband sonar in mobile robot navigation. In their method, they have used Doppler-shifting compensation successfully for navigation of mobile robots. Lin et al. [91] has discussed about topological navigation using ID Tag and WEB camera. They have discussed about node ID and post adjustment method for getting the direction angle and heading angle of the robot. They have stated that their method can be feasible for navigation of inter mobile robots.
Batalin et al. [92] have discussed about mobile robot navigation using sonar network. They have computed the direction of navigation within the sonar network using value iteration. They have done experimental verification to validate their proposed technique. Lee et al. [93] have discussed about a practical algorithm for topological navigation in corridor mapping of the mobile robot using sonar sensor. They have used circle following algorithm to handle obstacle avoidance situations during navigation.
Kim and Nak [94] have used the radio-frequency identification (RFID) system for navigation of mobile system. They have demonstrated that using the proposed system the robot can approach to a stationary target object. They have done various exercises to show the effectiveness of their proposed approach. Trajectory linearization based controller has been used by Liu et al. [95] to control the omnidirectional mobile robot. They have used sensor fusion method to combine onboard sensor and vision data system to provide accurate and reliable robot position and orientation during the navigation. Indoor navigation of a wheeled mobile robot has been discussed by Popa et al. [96]. The robot uses video images for navigational purpose. An odometric based OA algorithm has been implemented by them for navigation of the mobile robot. Ray et al. [97] have presented an integration of GPS and sonar based technique for the navigation of mobile robots. They have stated that, this mapping enables the robot to navigate among different GPS locations to get the longitude and latitude data for a particular location. They have implemented their technique in Pioneer robot for experimental verification.
Morkovic and Ivan [98] have studied on speaker localization and tracking with a microphone array on a mobile robot using von Mises distribution and particle filtering. They have used two most common microphone arrays to conduct the experiments in order to test their algorithms quantitatively and qualitatively.
Quintia et al. [99] have proposed a learning approach for mobile robot using reinforcement based strategy and a dynamic sensory data state mapping. They have claimed that a dynamic creation of state representation will help in mobile robot in controlling. Teimoori and Andrey [100] have discussed about navigation of a mobile robot using various sensory data. They have given mathematical analysis for path planning of mobile robots. They have shown computer simulation and experimental results using Pioneer robots.
Sanada et al. [101] have discussed about self-localization of an omnidirectional mobile robot using optical flow sensor. In their experimental method, the accuracy of self- location by dead reckoning and optical flow methods are evaluated using the motion capture methodology. They have claimed that the correct position can be achieved by optical flow sensor method rather than by the dead reckoning method. Delgado-Mata et al. [102] have discussed about the navigation of ethology inspired mobile robots. They have used ultrasound sensors for the navigation of mobile robots.
Espinace et al. [103] have discussed about indoor navigation of mobile robots using the adaptive object detection method. They have used various sensors to map the environment required during robot navigation. Siagian et al. [104] have discussed about hierarchical map representation for mobile robot localization and navigation. They have used several visual perception module such as place recognition, landmark recognition, and road lane detection for mapping the environment required for the robot navigation. Xu et al. [105] have discussed about sensor based robot navigation in complex environments. In their method, the motion detection of the robot is determined by using the key obstacle function segment sets. They have shown various simulation results to show the effectiveness of their proposed method. Pozna et al. [106] have discussed about pose estimation algorithm for the robot navigation. They have used various sensory data for navigation of the robot. They have shown the effectiveness of the proposed method using simulation results.