Multi-AGV global planning method based on network congestion model
A network congestion and global planning technology, applied in the direction of instruments, data processing applications, resources, etc., to achieve the effects of optimizing congestion waiting time, improving scheduling efficiency, and time performance advantages
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0056] The present invention provides a multi-AGV global planning method based on network congestion model, such as figure 1 As shown, it specifically includes the following steps:
[0057] 1) Establish a warehouse environment map and regionalize it, such as figure 2 As shown, the warehouse is divided into 16 areas, and the 4 areas on the right are set (Q 4 , Q 4 , Q 12 , Q 16 ) is the starting area, and other areas are the target areas. After the controller receives the task request, it obtains the number r and coordinates of each AGV in the storage environment, and assigns transportation tasks to each idle AGV, so as to determine the starting area O and target area E of each idle AGV's path in the current task, and set the starting area The initial area O is the parent node;
[0058] 2) Put the parent node into the close set, import the adjacent area of the parent node into the open set, if the target area is in the open set, then go to step 8), otherwise go to step...
Embodiment 2
[0090] In some storage systems, there are frequently moving obstacles other than AGV, and the environment is complex and changeable. Since the paths of frequently moving obstacles other than AGV are unpredictable, if based on the improved A* algorithm in Example 1, it still takes more secondary planning. In view of this situation, the A* algorithm in Embodiment 1 is changed to the D* algorithm, but the network congestion diffusion model is still used for improvement. Step 1) in the embodiment 1 no longer sets the starting area O as the parent node, and instead sets the target area E as the parent node; step 2) changes the parent node into the close set, and the adjacent parent node The area is imported into the open set. If the starting area is in the open set, go to step 8), otherwise go to step 3); when obstacles other than AGV enter the storage system, the improved D* algorithm can continue to use the previously calculated For the area path set between the target area and ...
Embodiment 3
[0093] In some storage systems, if certain areas are restricted by the surrounding environment, AGVs can only pass through a single channel, such as Figure 5 shown. AGVs can only move from area 6 to area 5, but AGVs in area 5 cannot move to area 6. In view of this situation, when using the network congestion diffusion model, the area set Ne connected to area 5 5 Although it is area 1, area 6 and area 9. But in computing the static and dynamic part A of the Langevin diffusion equation i (t) and B i (t), since some paths are single-channel, the formula is changed to the following form:
[0094]
[0095]
[0096] in, All AGVs in the inner area can move to area i. Taking area 9 as an example, its For area 5 and area 13. All areas within can be the target area of area i, taking area 5 as an example, its for zone 1 and zone 9.
[0097] Subsequent cost value G of region i i The calculation flow of (t) is consistent with Embodiment 1. Through the above improve...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More