Adaptive path planning for unknown environment monitoring
Nandhagopal Gomathi,Krishnamoorthi Rajathi
DOI: https://doi.org/10.3233/ais-220175
2023-08-02
Journal of Ambient Intelligence and Smart Environments
Abstract:The purpose of this paper is to offer a unique adaptive path planning framework to address a new challenge known as the Unknown environment Persistent Monitoring Problem (PMP). To identify the unknown events' occurrence location and likelihood, an unmanned ground vehicle (UGV) equipped with a Light Detection and Ranging (LIDAR) and camera is used to record such events in agriculture land. A certain level of detecting capability must be the distinct monitoring priority in order to keep track of them to a certain distance. First, to formulate a model, we developed an event-oriented modelling strategy for unknown environment perception and the effect is enumerated by uncertainty, which takes into account the sensor's detection capabilities, the detection interval, and monitoring weight. A mobile robot scheme utilizing LIDAR on integrative approach was created and experiments were carried out to solve the high equipment budget of Simultaneous Localization and Mapping (SLAM) for robotic systems. To map an unfamiliar location using the robotic operating system (ROS), the 3D visualization tool for Robot Operating System (RVIZ) was utilized, and GMapping software package was used for SLAM usage. The experimental results suggest that the mobile robot design pattern is viable to produce a high-precision map while lowering the cost of the mobile robot SLAM hardware. From a decision-making standpoint, we built a hybrid algorithm HSAStar (Hybrid SLAM & A Star) algorithm for path planning based on the event oriented modelling, allowing a UGV to continually monitor the perspectives of a path. The simulation results and analyses show that the proposed strategy is feasible and superior. The performance of the proposed hyb SLAM-A Star-APP method provides 34.95%, 27.38%, 33.21% and 29.68% lower execution time, 26.36%, 29.64% and 29.67% lower map duration compared with the existing methods, such as ACO-APF-APP, APFA-APP, GWO-APP and PSO-APP.
computer science, information systems,telecommunications, artificial intelligence