Research on preprocessing algorithm of indoor map partitioning and global path planning based on FAST DOI Creative Commons

Jifan Yang,

Xunding Pan,

Xiaoyang Liu

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: Nov. 30, 2023

Abstract Path planning is a critical factor in the successful performance of navigation tasks. This paper proposes novel approach for indoor map partitioning and global path-planning preprocessing. The proposed algorithm aims to enhance efficiency path tasks by eliminating irrelevant areas. In view deformation problem encountered original method, initially, contour detection employed identify eliminate obstacles. Subsequently, FAST utilized detect key points. These points are then subjected filtering clustering using K-means algorithm. Based on 8-neighborhood characteristics, door inflection within room selected. A retain points, which subsequently connected form line segments through averaging procedures. process ensures closure sub-room. Finally, domain function extract sub-room map, thereby completing process. centroid coordinate point data obtained from partitioning, two combinations used as starting end point, respectively, A* calculate store all information point. stored information, traversed areas, achieving preprocessing planning. simulation results showed that A*, Bi-A*, JPS, Dijkstra, PRM, RRT algorithms increased their rates 18.2%, 43.6%, 20.5%, 31.9%, 29.1%, 29.7%, respectively.

Language: Английский

A Survey of Machine Learning Approaches for Mobile Robot Control DOI Creative Commons
Monika Rybczak, Natalia Popowniak, Agnieszka Lazarowska

et al.

Robotics, Journal Year: 2024, Volume and Issue: 13(1), P. 12 - 12

Published: Jan. 9, 2024

Machine learning (ML) is a branch of artificial intelligence that has been developing at dynamic pace in recent years. ML also linked with Big Data, which are huge datasets need special tools and approaches to process them. algorithms make use data learn how perform specific tasks or appropriate decisions. This paper presents comprehensive survey have applied the task mobile robot control, they divided into following: supervised learning, unsupervised reinforcement learning. The distinction methods wheeled robots walking presented paper. strengths weaknesses compared formulated, future prospects proposed. results carried out literature review enable one state different tasks, such as position estimation, environment mapping, SLAM, terrain classification, obstacle avoidance, path following, walk, multirobot coordination. allowed us associate most commonly used robotic tasks. There still exist many open questions challenges complex limited computational resources on board robot; decision making motion control real time; adaptability changing environments; acquisition large volumes valuable data; assurance safety reliability robot’s operation. development for nature-inspired seems be challenging research issue there exists very amount solutions literature.

Language: Английский

Citations

13
A Path-Planning Approach for an Unmanned Vehicle in an Off-Road Environment Based on an Improved A* Algorithm DOI Creative Commons
Gaoyang Xie, Liqing Fang,

Xujun Su

et al.

World Electric Vehicle Journal, Journal Year: 2024, Volume and Issue: 15(6), P. 234 - 234

Published: May 29, 2024

Path planning for an unmanned vehicle in off-road uncertain environment is important navigation safety and efficiency. Regarding this, a global improved A* algorithm presented. Firstly, based on remote sensing images, the artificial potential field method used to describe distribution of risk environment, all types ground conditions are converted into travel time costs. Additionally, improvements include multi-directional node search algorithm, new line-of-sight designed which can sub-nodes more accurately, while factor passing-time cost added function. Finally, three kinds paths be calculated, including shortest path, path less risk, time-cost. The results simulation show that suitable vehicles complex environment. effectiveness verified by comparison between actual condition verification.

Language: Английский

Citations

3
APF-IBRRT*: A Global Path Planning Algorithm for Obstacle Avoidance Robots With Improved Iterative Search Efficiency DOI Creative Commons

Jiuyang Gao,

Xiang Zheng,

Pan Liu

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 124740 - 124750

Published: Jan. 1, 2024

Language: Английский

Citations

1
Algorithm for UAV path planning in high obstacle density environments: RFA-star DOI Creative Commons
Weijian Zhang, Jian Li, Weilin Yu

et al.

Frontiers in Plant Science, Journal Year: 2024, Volume and Issue: 15

Published: Oct. 17, 2024

Path planning is one of the key elements for achieving rapid and stable flight when unmanned aerial vehicles (UAVs) are conducting monitoring inspection tasks at ultra-low altitudes or in orchard environments. It involves finding optimal safe route between a given starting point target point. Achieving complex environments paramount. In characterized by high-density obstacles, stability UAVs remains focal research path algorithms. This study, utilizing feature attention mechanism, systematically identifies distinctive points on leading to development RFA-Star (R5DOS Feature Attention A-star) algorithm. MATLAB, random maps were generated assess performance The analysis focused evaluating effectiveness algorithm under varying obstacle density conditions different map sizes. Additionally, comparative analyses juxtaposed against three other Experimental results indicate that demonstrates shortest computation time, approximately 84%-94% faster than RJA-Star 51%-96% Improved A-Star. distance comparable algorithm, with slightly more searched nodes. Considering these factors collectively, exhibits relatively superior balance computational efficiency quality. consistently efficient across diverse However, comprehensive enhancement, further optimization necessary.

Language: Английский

Citations

1
Research on preprocessing algorithm of indoor map partitioning and global path planning based on FAST DOI Creative Commons

Jifan Yang,

Xunding Pan,

Xiaoyang Liu

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: Nov. 30, 2023

Abstract Path planning is a critical factor in the successful performance of navigation tasks. This paper proposes novel approach for indoor map partitioning and global path-planning preprocessing. The proposed algorithm aims to enhance efficiency path tasks by eliminating irrelevant areas. In view deformation problem encountered original method, initially, contour detection employed identify eliminate obstacles. Subsequently, FAST utilized detect key points. These points are then subjected filtering clustering using K-means algorithm. Based on 8-neighborhood characteristics, door inflection within room selected. A retain points, which subsequently connected form line segments through averaging procedures. process ensures closure sub-room. Finally, domain function extract sub-room map, thereby completing process. centroid coordinate point data obtained from partitioning, two combinations used as starting end point, respectively, A* calculate store all information point. stored information, traversed areas, achieving preprocessing planning. simulation results showed that A*, Bi-A*, JPS, Dijkstra, PRM, RRT algorithms increased their rates 18.2%, 43.6%, 20.5%, 31.9%, 29.1%, 29.7%, respectively.

Language: Английский

Citations

0