To solve the problem of energy consumption optimization of edge servers in the process of edge task unloading, we propose a task unloading algorithm based on reinforcement learning in this paper. The algorithm observes and analyzes the current environment state, selects the deployment location of edge tasks according to current states, and realizes the edge task unloading oriented to energy consumption optimization. To achieve the above goals, we first construct a network energy consumption model including servers' energy consumption and link transmission energy consumption, which improves the accuracy of network energy consumption evaluation. Because of the complexity and variability of the edge environment, this paper designs a task unloading algorithm based on Proximal Policy Optimization (PPO), besides we use Dijkstra to determine the connection path between edge servers where adjacent tasks are deployed. Finally, lots of simulation experiments verify the effectiveness of the proposed method in the process of task unloading. Compared with contrast algorithms, the average energy saving of the proposed algorithm can reach 22.69%.
Planning path rapidly and optimally is one of the key technologies for industrial manipulators. A novel method based on Memory-Goal-Biasing-Rapidly-exploring Random Tree is proposed to solve high-dimensional manipulation planning more rapidly and optimally. The tree extension of Memory-Goal-Biasing-Rapidly-exploring Random Tree can be divided into random extension and goal extension. In the goal extension, the nodes extended to the goal are recorded in a memory, and then the node closest to the goal is selected in the search tree excepting the nodes in the memory for overcoming the local minimum. In order to check collisions efficiently, the manipulator is simplified into several key points, and the obstacle area is appropriately enlarged for safety. Taking the redundant manipulator of Baxter robot as an example, the proposed algorithm is verified through MoveIt! software. The results show that Memory-Goal-Biasing-Rapidly-exploring Random Tree only takes a few seconds for the path planning of the redundant manipulator in some complex environments, and within an acceptable time, its optimization performance is better than that of traditional optimal method in terms of the obtained path costs and the corresponding standard deviation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.