Aiming at the dimension disaster problem, poor model generalization ability and deadlock problem in special obstacles environment caused by the increase of state information in the local path planning process of mobil...Aiming at the dimension disaster problem, poor model generalization ability and deadlock problem in special obstacles environment caused by the increase of state information in the local path planning process of mobile robot, this paper proposed a Double BP Q-learning algorithm based on the fusion of Double Q-learning algorithm and BP neural network. In order to solve the dimensional disaster problem, two BP neural network fitting value functions with the same network structure were used to replace the two <i>Q</i> value tables in Double Q-Learning algorithm to solve the problem that the <i>Q</i> value table cannot store excessive state information. By adding the mechanism of priority experience replay and using the parameter transfer to initialize the model parameters in different environments, it could accelerate the convergence rate of the algorithm, improve the learning efficiency and the generalization ability of the model. By designing specific action selection strategy in special environment, the deadlock state could be avoided and the mobile robot could reach the target point. Finally, the designed Double BP Q-learning algorithm was simulated and verified, and the probability of mobile robot reaching the target point in the parameter update process was compared with the Double Q-learning algorithm under the same condition of the planned path length. The results showed that the model trained by the improved Double BP Q-learning algorithm had a higher success rate in finding the optimal or sub-optimal path in the dense discrete environment, besides, it had stronger model generalization ability, fewer redundant sections, and could reach the target point without entering the deadlock zone in the special obstacles environment.展开更多
The forward-looking image sonar is a necessary vision device for Autonomous Underwater Vehicles （AUV）. Based on the acoustic image received from forward-looking image sonar, AUV local path is planned. When the envir...The forward-looking image sonar is a necessary vision device for Autonomous Underwater Vehicles （AUV）. Based on the acoustic image received from forward-looking image sonar, AUV local path is planned. When the environment model is made to adapt to local path planning, an iterative algorithm of binary conversion is used for image segmentation. Raw data of the acoustic image, which were received from serial port, are processed. By the use of ＂Mathematic Morphology＂ to filter noise, a mathematic model of environment for local path planning is established after coordinate transformation. The optimal path is searched by the distant transmission （Dt） algorithm. Simulation is conducted for the analysis of the algorithm. Experiment on the sea proves it reliable.展开更多
Conducting hydrodynamic and physical motion simulation tests using a large-scale self-propelled model under actual wave conditions is an important means for researching environmental adaptability of ships. During the ...Conducting hydrodynamic and physical motion simulation tests using a large-scale self-propelled model under actual wave conditions is an important means for researching environmental adaptability of ships. During the navigation test of the self-propelled model, the complex environment including various port facilities, navigation facilities, and the ships nearby must be considered carefully, because in this dense environment the impact of sea waves and winds on the model is particularly significant. In order to improve the security of the self-propelled model, this paper introduces the Q learning based on reinforcement learning combined with chaotic ideas for the model＇s collision avoidance, in order to improve the reliability of the local path planning. Simulation and sea test results show that this algorithm is a better solution for collision avoidance of the self navigation model under the interference of sea winds and waves with good adaptability.展开更多
For mobile anchor node static path planning cannot accord the actual distribution of node for dynamic adjustment. We take advantage of the high localization accuracy and low computational complexity of ad-hoc localiza...For mobile anchor node static path planning cannot accord the actual distribution of node for dynamic adjustment. We take advantage of the high localization accuracy and low computational complexity of ad-hoc localization system( AHLos)algorithm. This article introduces mobile anchor nodes instead of the traditional fixed anchor nodes to improve the algorithm. The result shows that, through introduce the mobile anchor node, the information of initial anchor nodes can be configured more flexible.Meanwhile,with the use of the approximate location and the transition path,the distance and energy consumption of the mobile anchor node is greatly reduced.展开更多
This paper presents a development of a novel path planning algorithm, called Generalized Laser simulator (GLS), for solving the mobilerobot path planning problem in a two-dimensional map with the presence ofconstraint...This paper presents a development of a novel path planning algorithm, called Generalized Laser simulator (GLS), for solving the mobilerobot path planning problem in a two-dimensional map with the presence ofconstraints. This approach gives the possibility to find the path for a wheelmobile robot considering some constraints during the robot movement inboth known and unknown environments. The feasible path is determinedbetween the start and goal positions by generating wave of points in alldirection towards the goal point with adhering to constraints. In simulation,the proposed method has been tested in several working environments withdifferent degrees of complexity. The results demonstrated that the proposedmethod is able to generate efficiently an optimal collision-free path. Moreover,the performance of the proposed method was compared with the A-star andlaser simulator (LS) algorithms in terms of path length, computational timeand path smoothness. The results revealed that the proposed method hasshortest path length, less computational time and the best smooth path. Asan average, GLS is faster than A∗ and LS by 7.8 and 5.5 times, respectivelyand presents a path shorter than A∗ and LS by 1.2 and 1.5 times. In orderto verify the performance of the developed method in dealing with constraints, an experimental study was carried out using a Wheeled Mobile Robot(WMR) platform in labs and roads. The experimental work investigates acomplete autonomous WMR path planning in the lab and road environmentsusing a live video streaming. Local maps were built using data from a live video streaming with real-time image processing to detect segments of theanalogous-road in lab or real-road environments. The study shows that theproposed method is able to generate shortest path and best smooth trajectoryfrom start to goal points in comparison with laser simulator.展开更多
针对AGV(Automated Guided Vehicle)叉车处于环境信息未知或环境动态变化情况下的路径规划及导航问题,文中提出了一种由YOLOv5(You Only Look Once version 5)目标检测算法获取目标位置。根据目标位置规划出全局基础路径,再融合DWA(Dyna...针对AGV(Automated Guided Vehicle)叉车处于环境信息未知或环境动态变化情况下的路径规划及导航问题,文中提出了一种由YOLOv5(You Only Look Once version 5)目标检测算法获取目标位置。根据目标位置规划出全局基础路径,再融合DWA(Dynamic Window Approach)局部动态路径规划算法进行AGV路径规划与导航,使AGV叉车在未知环境或局部环境信息未知的环境中能快速识别目标位置并完成路径规划到达目标位置。实验结果表明,相较于改进前方法,文中所提方法在路径长度、耗费时间以及AGV叉车航向误差方面均有良好表现,路径长度平均减少12%,耗费时间平均减少约5%且AGV航向与目标航向的平均误差在5°以内。所提方法提高了AGV叉车在未知环境中的工作效率以及工作灵活性。展开更多
文摘Aiming at the dimension disaster problem, poor model generalization ability and deadlock problem in special obstacles environment caused by the increase of state information in the local path planning process of mobile robot, this paper proposed a Double BP Q-learning algorithm based on the fusion of Double Q-learning algorithm and BP neural network. In order to solve the dimensional disaster problem, two BP neural network fitting value functions with the same network structure were used to replace the two <i>Q</i> value tables in Double Q-Learning algorithm to solve the problem that the <i>Q</i> value table cannot store excessive state information. By adding the mechanism of priority experience replay and using the parameter transfer to initialize the model parameters in different environments, it could accelerate the convergence rate of the algorithm, improve the learning efficiency and the generalization ability of the model. By designing specific action selection strategy in special environment, the deadlock state could be avoided and the mobile robot could reach the target point. Finally, the designed Double BP Q-learning algorithm was simulated and verified, and the probability of mobile robot reaching the target point in the parameter update process was compared with the Double Q-learning algorithm under the same condition of the planned path length. The results showed that the model trained by the improved Double BP Q-learning algorithm had a higher success rate in finding the optimal or sub-optimal path in the dense discrete environment, besides, it had stronger model generalization ability, fewer redundant sections, and could reach the target point without entering the deadlock zone in the special obstacles environment.
文摘The forward-looking image sonar is a necessary vision device for Autonomous Underwater Vehicles （AUV）. Based on the acoustic image received from forward-looking image sonar, AUV local path is planned. When the environment model is made to adapt to local path planning, an iterative algorithm of binary conversion is used for image segmentation. Raw data of the acoustic image, which were received from serial port, are processed. By the use of ＂Mathematic Morphology＂ to filter noise, a mathematic model of environment for local path planning is established after coordinate transformation. The optimal path is searched by the distant transmission （Dt） algorithm. Simulation is conducted for the analysis of the algorithm. Experiment on the sea proves it reliable.
基金Foundation item： Supported by the National Natural Science Foundation of China under Grant No.61100005.
文摘Conducting hydrodynamic and physical motion simulation tests using a large-scale self-propelled model under actual wave conditions is an important means for researching environmental adaptability of ships. During the navigation test of the self-propelled model, the complex environment including various port facilities, navigation facilities, and the ships nearby must be considered carefully, because in this dense environment the impact of sea waves and winds on the model is particularly significant. In order to improve the security of the self-propelled model, this paper introduces the Q learning based on reinforcement learning combined with chaotic ideas for the model＇s collision avoidance, in order to improve the reliability of the local path planning. Simulation and sea test results show that this algorithm is a better solution for collision avoidance of the self navigation model under the interference of sea winds and waves with good adaptability.
基金National Natural Science Foundations of China(Nos.U1162202,61203157)the Fundamental Research Funds for the Central Universities and Shanghai Leading Academic Discipline Project,China(No.B504)
文摘For mobile anchor node static path planning cannot accord the actual distribution of node for dynamic adjustment. We take advantage of the high localization accuracy and low computational complexity of ad-hoc localization system( AHLos)algorithm. This article introduces mobile anchor nodes instead of the traditional fixed anchor nodes to improve the algorithm. The result shows that, through introduce the mobile anchor node, the information of initial anchor nodes can be configured more flexible.Meanwhile,with the use of the approximate location and the transition path,the distance and energy consumption of the mobile anchor node is greatly reduced.
基金The authors would like to thank the United Arab Emirates University for funding this work under Start-Up grant[G00003321].
文摘This paper presents a development of a novel path planning algorithm, called Generalized Laser simulator (GLS), for solving the mobilerobot path planning problem in a two-dimensional map with the presence ofconstraints. This approach gives the possibility to find the path for a wheelmobile robot considering some constraints during the robot movement inboth known and unknown environments. The feasible path is determinedbetween the start and goal positions by generating wave of points in alldirection towards the goal point with adhering to constraints. In simulation,the proposed method has been tested in several working environments withdifferent degrees of complexity. The results demonstrated that the proposedmethod is able to generate efficiently an optimal collision-free path. Moreover,the performance of the proposed method was compared with the A-star andlaser simulator (LS) algorithms in terms of path length, computational timeand path smoothness. The results revealed that the proposed method hasshortest path length, less computational time and the best smooth path. Asan average, GLS is faster than A∗ and LS by 7.8 and 5.5 times, respectivelyand presents a path shorter than A∗ and LS by 1.2 and 1.5 times. In orderto verify the performance of the developed method in dealing with constraints, an experimental study was carried out using a Wheeled Mobile Robot(WMR) platform in labs and roads. The experimental work investigates acomplete autonomous WMR path planning in the lab and road environmentsusing a live video streaming. Local maps were built using data from a live video streaming with real-time image processing to detect segments of theanalogous-road in lab or real-road environments. The study shows that theproposed method is able to generate shortest path and best smooth trajectoryfrom start to goal points in comparison with laser simulator.
文摘针对AGV(Automated Guided Vehicle)叉车处于环境信息未知或环境动态变化情况下的路径规划及导航问题,文中提出了一种由YOLOv5(You Only Look Once version 5)目标检测算法获取目标位置。根据目标位置规划出全局基础路径,再融合DWA(Dynamic Window Approach)局部动态路径规划算法进行AGV路径规划与导航,使AGV叉车在未知环境或局部环境信息未知的环境中能快速识别目标位置并完成路径规划到达目标位置。实验结果表明,相较于改进前方法,文中所提方法在路径长度、耗费时间以及AGV叉车航向误差方面均有良好表现,路径长度平均减少12%,耗费时间平均减少约5%且AGV航向与目标航向的平均误差在5°以内。所提方法提高了AGV叉车在未知环境中的工作效率以及工作灵活性。
