By integrating deep neural networks with reinforcement learning,the Double Deep Q Network(DDQN)algorithm overcomes the limitations of Q-learning in handling continuous spaces and is widely applied in the path planning...By integrating deep neural networks with reinforcement learning,the Double Deep Q Network(DDQN)algorithm overcomes the limitations of Q-learning in handling continuous spaces and is widely applied in the path planning of mobile robots.However,the traditional DDQN algorithm suffers from sparse rewards and inefficient utilization of high-quality data.Targeting those problems,an improved DDQN algorithm based on average Q-value estimation and reward redistribution was proposed.First,to enhance the precision of the target Q-value,the average of multiple previously learned Q-values from the target Q network is used to replace the single Q-value from the current target Q network.Next,a reward redistribution mechanism is designed to overcome the sparse reward problem by adjusting the final reward of each action using the round reward from trajectory information.Additionally,a reward-prioritized experience selection method is introduced,which ranks experience samples according to reward values to ensure frequent utilization of high-quality data.Finally,simulation experiments are conducted to verify the effectiveness of the proposed algorithm in fixed-position scenario and random environments.The experimental results show that compared to the traditional DDQN algorithm,the proposed algorithm achieves shorter average running time,higher average return and fewer average steps.The performance of the proposed algorithm is improved by 11.43%in the fixed scenario and 8.33%in random environments.It not only plans economic and safe paths but also significantly improves efficiency and generalization in path planning,making it suitable for widespread application in autonomous navigation and industrial automation.展开更多
Periodic motion planning for an under-actuated system is rather difficult due to differential dynamic constraints imposed by passive dynamics, and it becomes more difficult for a system with higher underactuation degr...Periodic motion planning for an under-actuated system is rather difficult due to differential dynamic constraints imposed by passive dynamics, and it becomes more difficult for a system with higher underactuation degree, that is with a higher difference between the number of degrees of freedom and the number of independent control inputs. However, from another point of view, these constraints also mean some relation between state variables and could be used in the motion planning.We consider a double rotary pendulum, which has an underactuation degree 2. A novel periodic motion planning is presented based on an optimization search. A necessary condition for existence of the whole periodic trajectory is given because of the higher underactuation degree of the system. Moreover this condition is given to make virtual holonomic constraint(VHC) based control design feasible. Therefore, an initial guess for the optimization of planning a feasible periodic motion is based on this necessary condition. Then, VHCs are used for the system transformation and transverse linearization is used to design a static state feedback controller with periodic matrix function gain. The controller gain is found through another optimization procedure. The effectiveness of initial guess and performance of the closed-loop system are illustrated through numerical simulations.展开更多
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
Training talents for the society is the responsibility of colleges and universities.The society needs applied and innovative art design majors.In order to cultivate talents needed by society and keep up with the devel...Training talents for the society is the responsibility of colleges and universities.The society needs applied and innovative art design majors.In order to cultivate talents needed by society and keep up with the development plan of the Ministry of Education,higher vocational colleges need to reform.This paper adopts the method of theoretical analysis to elaborate from the four aspects of focusing equally on science and education,promote learning by competition;integrating industry and education,nurturing talents together;keeping the mission in mind while serving students;and finding the right positioning,giving full play to the advantages.展开更多
This article analyzed the ten strategic pillar industry clusters and the ten strategic emerging industry clusters which were confirmed and announced by Guangdong Province in May 2020 in addition to the 14 professional...This article analyzed the ten strategic pillar industry clusters and the ten strategic emerging industry clusters which were confirmed and announced by Guangdong Province in May 2020 in addition to the 14 professional groups of schools in Guangdong Province which were shortlisted in the"Double High Plan."With comparison and analysis,the degree of coincidence between the professional groups of higher vocational colleges and strategic industrial clusters in Guangdong Province which covered most of the industrial clusters was relatively high.The higher vocational colleges that were shortlisted in the"Double High Plan"still required dynamic adjustments in the construction of professional groups to achieve an allrounded integration of the supply of talents and industrial demands.展开更多
In order to elucidate the relationship between talent recruitment and teacher allocation management in higher vocational colleges under the background of the"Double High Plan",this paper studies and analyzes...In order to elucidate the relationship between talent recruitment and teacher allocation management in higher vocational colleges under the background of the"Double High Plan",this paper studies and analyzes the current status of talent recruitment in Chinese higher vocational colleges in conjunction with the characteristics of talent resource allocation in higher vocational colleges and proposes to elucidate the priorities and improve supporting policies with reasonable planning to improve the optimization guideline of talent recruitment efficiency,so as to better accomplish the building of the national"double high plan"and improve the ability of higher vocational colleges to nurture more outstanding talents for the society.展开更多
Starting from the background of the"Double High Plan"in conjunction with the role of personnel file management in advancing the"Double High Plan",this paper analyzes the current status of personnel...Starting from the background of the"Double High Plan"in conjunction with the role of personnel file management in advancing the"Double High Plan",this paper analyzes the current status of personnel file management in higher vocational colleges and the necessity of personnel management in higher vocational colleges,and explores the effective measures to strengthen the informatization of personnel files management to make it more reasonable,standardized and informatized.展开更多
In order to cultivate more excellent talents in art and design majors following the requirements of the"13th Five-Year Plan"national education development,this paper analyzes in-depth on the current status o...In order to cultivate more excellent talents in art and design majors following the requirements of the"13th Five-Year Plan"national education development,this paper analyzes in-depth on the current status of higher vocational education in art design profession under the background of the"Double High"plan and the challenges faced,and proposes the building of school-enterprise"Double Subject"education system and the establishment of a"diversified"evaluation system.Deepening the integration of industry and education,schools and enterprises jointly explore modern apprenticeship talent training models for innovative art and design majors,and provide a strong guarantee for the implementation of the"Double High"plan modern apprenticeship talent training model.展开更多
基金funded by National Natural Science Foundation of China(No.62063006)Guangxi Science and Technology Major Program(No.2022AA05002)+1 种基金Key Laboratory of AI and Information Processing(Hechi University),Education Department of Guangxi Zhuang Autonomous Region(No.2022GXZDSY003)Central Leading Local Science and Technology Development Fund Project of Wuzhou(No.202201001).
文摘By integrating deep neural networks with reinforcement learning,the Double Deep Q Network(DDQN)algorithm overcomes the limitations of Q-learning in handling continuous spaces and is widely applied in the path planning of mobile robots.However,the traditional DDQN algorithm suffers from sparse rewards and inefficient utilization of high-quality data.Targeting those problems,an improved DDQN algorithm based on average Q-value estimation and reward redistribution was proposed.First,to enhance the precision of the target Q-value,the average of multiple previously learned Q-values from the target Q network is used to replace the single Q-value from the current target Q network.Next,a reward redistribution mechanism is designed to overcome the sparse reward problem by adjusting the final reward of each action using the round reward from trajectory information.Additionally,a reward-prioritized experience selection method is introduced,which ranks experience samples according to reward values to ensure frequent utilization of high-quality data.Finally,simulation experiments are conducted to verify the effectiveness of the proposed algorithm in fixed-position scenario and random environments.The experimental results show that compared to the traditional DDQN algorithm,the proposed algorithm achieves shorter average running time,higher average return and fewer average steps.The performance of the proposed algorithm is improved by 11.43%in the fixed scenario and 8.33%in random environments.It not only plans economic and safe paths but also significantly improves efficiency and generalization in path planning,making it suitable for widespread application in autonomous navigation and industrial automation.
基金supported by China Scholarship Council (201504980073) for Zeguo Wang to visit Umea University
文摘Periodic motion planning for an under-actuated system is rather difficult due to differential dynamic constraints imposed by passive dynamics, and it becomes more difficult for a system with higher underactuation degree, that is with a higher difference between the number of degrees of freedom and the number of independent control inputs. However, from another point of view, these constraints also mean some relation between state variables and could be used in the motion planning.We consider a double rotary pendulum, which has an underactuation degree 2. A novel periodic motion planning is presented based on an optimization search. A necessary condition for existence of the whole periodic trajectory is given because of the higher underactuation degree of the system. Moreover this condition is given to make virtual holonomic constraint(VHC) based control design feasible. Therefore, an initial guess for the optimization of planning a feasible periodic motion is based on this necessary condition. Then, VHCs are used for the system transformation and transverse linearization is used to design a static state feedback controller with periodic matrix function gain. The controller gain is found through another optimization procedure. The effectiveness of initial guess and performance of the closed-loop system are illustrated through numerical simulations.
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Training talents for the society is the responsibility of colleges and universities.The society needs applied and innovative art design majors.In order to cultivate talents needed by society and keep up with the development plan of the Ministry of Education,higher vocational colleges need to reform.This paper adopts the method of theoretical analysis to elaborate from the four aspects of focusing equally on science and education,promote learning by competition;integrating industry and education,nurturing talents together;keeping the mission in mind while serving students;and finding the right positioning,giving full play to the advantages.
文摘This article analyzed the ten strategic pillar industry clusters and the ten strategic emerging industry clusters which were confirmed and announced by Guangdong Province in May 2020 in addition to the 14 professional groups of schools in Guangdong Province which were shortlisted in the"Double High Plan."With comparison and analysis,the degree of coincidence between the professional groups of higher vocational colleges and strategic industrial clusters in Guangdong Province which covered most of the industrial clusters was relatively high.The higher vocational colleges that were shortlisted in the"Double High Plan"still required dynamic adjustments in the construction of professional groups to achieve an allrounded integration of the supply of talents and industrial demands.
文摘In order to elucidate the relationship between talent recruitment and teacher allocation management in higher vocational colleges under the background of the"Double High Plan",this paper studies and analyzes the current status of talent recruitment in Chinese higher vocational colleges in conjunction with the characteristics of talent resource allocation in higher vocational colleges and proposes to elucidate the priorities and improve supporting policies with reasonable planning to improve the optimization guideline of talent recruitment efficiency,so as to better accomplish the building of the national"double high plan"and improve the ability of higher vocational colleges to nurture more outstanding talents for the society.
文摘Starting from the background of the"Double High Plan"in conjunction with the role of personnel file management in advancing the"Double High Plan",this paper analyzes the current status of personnel file management in higher vocational colleges and the necessity of personnel management in higher vocational colleges,and explores the effective measures to strengthen the informatization of personnel files management to make it more reasonable,standardized and informatized.
文摘In order to cultivate more excellent talents in art and design majors following the requirements of the"13th Five-Year Plan"national education development,this paper analyzes in-depth on the current status of higher vocational education in art design profession under the background of the"Double High"plan and the challenges faced,and proposes the building of school-enterprise"Double Subject"education system and the establishment of a"diversified"evaluation system.Deepening the integration of industry and education,schools and enterprises jointly explore modern apprenticeship talent training models for innovative art and design majors,and provide a strong guarantee for the implementation of the"Double High"plan modern apprenticeship talent training model.
