Mobile Robot Path Planning Based on Kinematically Constrained A-Star Algorithm and DWA Fusion Algorithm

Liu, Yanjie; Wang, Chao; Wu, Heng; Wei, Yanlong

doi:10.3390/math11214552

Cited by 13 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The existing path planning methods can be divided into global path planning [26] and local path planning [27] . Commonly used algorithms include ant colony algorithm [28] , A* algorithm [29] , random number algorithm [30] and greedy algorithm, etc. In this paper, five evaluation indexes that are more important in the medical environment are used as the scoring criteria for the algorithm radar map.…”

Section: Selection Of Path Planning Algorithmmentioning

confidence: 99%

Obstacle avoidance strategy and path planning of medical AGV based on the bionic characteristics of antelope migration

Hu,

Niu,

Gao

et al. 2024

Preprint

View full text Add to dashboard Cite

Aiming at the problem that Automated guided vehicle (AGV) faces dynamic and static obstacles in the process of transporting patients in medical environment and needs to avoid in real time, inspired by the behavior of antelopes during migration, this paper proposes a bionic obstacle avoidance strategy based on adaptive behavior of antelopes. The traditional artificial potential field and dynamic window algorithm are improved by using the bionic characteristics of antelope migration.By adding new potential field force points and improving the window size, the success rate and prediction range of AGV navigation are improved.Simulation experiments were carried out through the numerical simulation platform, and the verification results showed that:The bionic obstacle avoidance strategy proposed in this paper can avoid dynamic and static obstacles at the same time. In the example, the success rate of path planning is increased by 34%, the running time is reduced by 33%, and the average path length is reduced by 1%. The proposed method can realize the integration of “dynamic and static” avoidance in the process of transporting patients, and effectively save the time of transporting patients by AGV.It provides a theoretical basis for realizing obstacle avoidance and rapid loading of AGV in medical environment.

show abstract

Section: Selection Of Path Planning Algorithmmentioning

confidence: 99%

Obstacle avoidance strategy and path planning of medical AGV based on the bionic characteristics of antelope migration

Hu,

Niu,

Gao

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Based on this, the algorithm needs to convert the sparse critical path points into denser equidistant path points. In this paper, key point densification is based on the B-spline curve [22][23][24][25][26].…”

Section: Key Point Density Strategymentioning

confidence: 99%

Research on fusing improved A* and adaptive DWA path planning algorithms

LIAO,

WANG,

QIN

2024

Preprint

View full text Add to dashboard Cite

To satisfy the performance requirements of robot path planning, an algorithm combining the improved A* algorithm and the improved Dynamic Window Approach (DWA) is proposed, which results in shorter path lengths, improved search efficiency, and path smoothness. Aiming at the challenges of the traditional A* algorithm in complex scenarios, a new heuristic function based on Manhattan and diagonal is designed, and then weights are dynamically assigned to obtain the global shortest path and the least search time. Then, an improved search strategy based on 8-neighborhoods is proposed, which improves the search efficiency and reduces the time consumption of the traditional 8-neighborhood 8-direction search method by dynamically assigning the optimal search direction of the current node. On the other hand, the traditional DWA algorithm faces some challenges, such as the paths are not globally optimal, the path planning may fail or path length may increase, the azimuthal coefficient is rigid, and the algorithm is computationally intensive. For these problems, a keypoint densification strategy is proposed to modify the deflected paths, adaptively adjust the azimuth function coefficients, and limit the range of the obstacle distance function. Finally, the proposed improved A* algorithm and fusion algorithm are compared with the existing methods. The simulation results under the ROS system show that the improved A* algorithm can generate the shortest global path in complex environments, the average path length is reduced by 3.95%, and the average path searching time is shortened by 21.62%. For the fused algorithm, the average path length and the average runtime are reduced by 5.95% and 8.7% in the moving obstacles environment.

show abstract

“…In addition, utilizing actions close to the target to train the action policy results in a better way for the robot to reach the target point and in a better overall action effect of the agent, and it also improves the sampling efficiency of the training. When the robot is in a dangerous area, the robot is controlled to "extrication" using a fast path-generation algorithm, which is provided directly by the ROS (Robot Operating System) navigation package and is a fusion of the traditional A* and DWA algorithms [21], so it will not be introduced here. The data generated by the above fast path-generation algorithm are recorded as expert data, and the blue trajectory in Figure 1 is the expert trajectory for "extrication", and when it is out of the danger zone, the intervention is ended and is changed to the agent method to control the robot movement.…”

Section: Sac Reinforcement-learning Navigation Policy Incorporating E...mentioning

confidence: 99%

“…Remote Sens. 2024, 16,2072 4 of 29 by the ROS (Robot Operating System) navigation package and is a fusion of the traditional A* and DWA algorithms [21], so it will not be introduced here. The data generated by the above fast path-generation algorithm are recorded as expert data, and the blue trajectory in Figure 1 is the expert trajectory for "extrication", and when it is out of the danger zone, the intervention is ended and is changed to the agent method to control the robot movement.…”

Section: Sac Reinforcement-learning Navigation Policy Incorporating E...mentioning

confidence: 99%

A Soft Actor-Critic Deep Reinforcement-Learning-Based Robot Navigation Method Using LiDAR

Liu,

Wang,

Zhao

et al. 2024

Remote Sensing

Self Cite

View full text Add to dashboard Cite

When there are dynamic obstacles in the environment, it is difficult for traditional path-generation algorithms to achieve desired obstacle-avoidance results. To solve this problem, we propose a robot navigation control method based on SAC (Soft Actor-Critic) Deep Reinforcement Learning. Firstly, we use a fast path-generation algorithm to control the robot to generate expert trajectories when the robot encounters danger as well as when it approaches a target, and we combine SAC reinforcement learning with imitation learning based on expert trajectories to improve the safety of training. Then, for the hybrid data consisting of agent data and expert data, we use an improved prioritized experience replay method to improve the learning efficiency of the policies. Finally, we introduce RNN (Recurrent Neural Network) units into the network structure of the SAC Deep Reinforcement-Learning navigation policy to improve the agent’s transfer inference ability in a new environment and obstacle-avoidance ability in dynamic environments. Through simulation and practical experiments, it is fully verified that our method has a higher training efficiency and navigation success rate compared to state-of-the-art reinforcement-learning algorithms, which further enhances the obstacle-avoidance capability of the robot system.

show abstract

Mobile Robot Path Planning Based on Kinematically Constrained A-Star Algorithm and DWA Fusion Algorithm

Cited by 13 publications

References 35 publications

Obstacle avoidance strategy and path planning of medical AGV based on the bionic characteristics of antelope migration

Obstacle avoidance strategy and path planning of medical AGV based on the bionic characteristics of antelope migration

Research on fusing improved A* and adaptive DWA path planning algorithms

A Soft Actor-Critic Deep Reinforcement-Learning-Based Robot Navigation Method Using LiDAR

Contact Info

Product

Resources

About