z-logo
open-access-imgOpen Access
Study of Goal Oriented Obstacle Avoidance for Mobile Robot
Author(s) -
Yeon Taek Oh
Publication year - 2021
Publication title -
international journal emerging technology and advanced engineering
Language(s) - English
Resource type - Journals
ISSN - 2250-2459
DOI - 10.46338/ijetae1021_02
Subject(s) - obstacle avoidance , mobile robot , path (computing) , obstacle , position (finance) , robot , computer science , control theory (sociology) , motion planning , controller (irrigation) , mathematical optimization , shortest path problem , simulation , artificial intelligence , mathematics , control (management) , graph , agronomy , finance , theoretical computer science , political science , law , economics , biology , programming language
—This paper suggests simultaneous mechanism which makes UGV to secure its safety and recreate optimum path right after obstacle avoidance occurs. Research target and sensors is shown and the suggested optimum region path algorithm is presented. Also, this paper provides the performance of suggested algorithm by simulation, and will propose the direction of the future plans. The fixed and varying weight algorithm satisfy the constraints, once the robot arrives at the goal position in the simulation. However, the fixed weight algorithm was not able to drive on the shortest path, which is optimal, with many other successful weight values. On the other hand, the varying weight algorithm successfully generated the optimized path by changing the weight values in term of local environment. Especially, in the second simulation, the robot was able to safely arrive to the goal with little time period. This paper proposed VTV and Local Optimal path algorithm that the robot can generate the optimal path to the goal position without colliding obstacles as adapting Fuzzy controller which continuously optimizes the weight values of the cost function in terms of local environment

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here