Wafer extraction path optimization method based on adaptive heuristic and hierarchical search

Abstract

The invention is suitable for the technical field of automation equipment control and path planning, and provides a wafer extraction path optimization method based on adaptive heuristic and hierarchical search, comprising the following steps: Step 1, initialization; step 2, heuristic value calculation: calculating an adaptive heuristic function value according to the current node to guide search; step 3, judging whether an end point is reached or not; step 4, path cost calculation is carried out; step 5, judging node expansion: judging whether to expand a current node according to a self-adaptive expansion condition, and preferentially expanding a node which is most likely to approach an end point; step 6, layered search of paths: layered search is carried out gradually from the outer layer to the inner layer, independent path planning is carried out on each layer, and the search range is reduced; step 7, path backtracking is carried out; step 8, updating a search area; step 9, continuing or terminating the search; and step 10, outputting a result. On the premise of ensuring the shortest path, the start-stop frequency of the motor is reduced, the operation efficiency of automatic equipment is improved, the service life of the equipment is prolonged, and the energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of automated equipment control and path planning, and in particular to a wafer extraction path optimization method based on adaptive heuristics and hierarchical search. Background Art

[0002] Wafer Handling or Wafer Extraction Equipment is one of the key equipment used for automatic handling and extraction of wafers in the semiconductor manufacturing process. These devices are usually used for wafer transmission, removal, transfer and placement on the production line. They are efficient, accurate and stable, and are necessary tools for achieving efficient production in the semiconductor industry.

[0003] When performing extraction tasks, wafer automated extraction equipment usually needs to consider how to find the shortest path in a complex path while reducing the frequent start and stop operations of the equipment. This not only affects the efficiency of the system, but also increases the wear and tear of the equipment and energy consumption.

[0004] Existing path planning algorithms, such as the traditional A* algorithm, can find the optimal path, but their performance is often not ideal when encountering complex environments, especially failing to fully optimize the number of motor starts and stops, resulting in reduced production efficiency. Summary of the invention

[0005] The purpose of the present invention is to provide a wafer extraction path optimization method based on adaptive heuristics and hierarchical search to solve the problems in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a wafer extraction path optimization method based on adaptive heuristics and hierarchical search, comprising the following steps:

[0007] Step 1, initialization:

[0008] Set the starting point and end point, define the grid environment and initial parameters;

[0009] Step 2, heuristic value calculation:

[0010] Calculate an adaptive heuristic function value based on the current node to guide the search;

[0011] Step 3, determine whether it has reached the end point:

[0012] If the current node is the end point, jump to Step 10, stop and output the path, otherwise continue;

[0013] Step 4, path cost calculation:

[0014] Calculate the moving cost from the current node to the next node according to the path cost formula;

[0015] Step 5, determine the node expansion:

[0016] Determine whether to expand the current node based on the adaptive expansion condition, and give priority to expanding the node that is most likely to be close to the end point;

[0017] Step 6, hierarchical search path:

[0018] Gradually conduct hierarchical search from the outer layer to the inner layer, and perform independent path planning at each layer to reduce the search scope;

[0019] Step 7, path backtracking:

[0020] If the search at a certain level fails or is invalid, return to Step 5, reselect the previous level node, and return to the previous level search;

[0021] Step 8, update the search area:

[0022] Dynamically adjust the expansion area, adjust the search parameters according to the results, and proceed to the next step;

[0023] Step 9, continue or terminate the search:

[0024] Determine whether the search needs to continue. If there is no valid path, return to Step 4. If the path is valid, output the final path and complete the planning.

[0025] Step 10, result output:

[0026] The path planning results are output to guide the equipment to extract the wafer along the optimized path.

[0027] Preferably, the calculation formula of the adaptive heuristic function in Step 2 is as follows:

[0028] h′(n)=h(n)+λ1D(n)+λ2H(n);

[0029] Among them, h′(n) is the adaptive heuristic value used to guide the search process;

[0030] Where h(n) is the traditional heuristic value.

[0031] Preferably, h(n) is usually calculated using the Euclidean distance, and the calculation formula is as follows:

[0032]

[0033] Among them, x goal and goal is the coordinate of the target node;

[0034] Among them, x n and n is the coordinate of the current node;

[0035] Among them, λ1 and λ2 are adjustment parameters used to control the influence of D(n) and H(n) on the adaptive heuristic value; D(n) is the local environment density adjustment item, which reflects the density of obstacles around the current node;

[0036] H(n) is the historical path correction term, which is used to give priority to valid path directions that have been used.

[0037] Preferably, the calculation formula of D(n) is as follows:

[0038]

[0039] Among them, α is the adjustment parameter, which controls the influence of density on the heuristic function;

[0040] Among them, d good (n) is the distance from the current node to the n nearest good chips;

[0041] Among them, d bad (n) is the distance from the current node n to the nearest obstacle;

[0042] The calculation formula of H(n) is as follows:

[0043]

[0044] Among them, P history is the node set in the historical path;

[0045] Among them, dist(n,p) is the distance between the current node n and the historical node p.

[0046] Preferably, the hierarchical search path divides the entire search area into several layers, and each layer is searched independently according to its complexity. The specific steps are as follows:

[0047] First, search the outer layer and then gradually move to the inner layer. The search of each layer can reduce the search space. Through layered search, the scope of each search can be reduced and path planning can be accelerated.

[0048] The calculation formula of the hierarchical path cost function is as follows:

[0049]

[0050] Among them, C total is the total path cost, which represents the total cost from the starting position to the target node;

[0051] Among them, Clayer (S i ) is the path cost of the i-th layer, which consists of the actual cost and the heuristic cost.

[0052] Preferably, the C layer (S i ) is calculated as follows:

[0053] C layer (S i )=g(S i )+h(S i );

[0054] Where g(Si) is the distance from the starting position to layer S i The cumulative cost represents the actual cost on the path;

[0055] Among them, h(S i ) is a heuristic estimate within the layer, usually using Manhattan distance or Euclidean distance.

[0056] Preferably, the g(S i ) is calculated as follows:

[0057]

[0058] Among them, cost(n,n prev ) is the node n and the previous node n prev The actual cost between.

[0059] Preferably, the mechanism of adaptive node expansion is to dynamically adjust the priority of the node to be expanded according to the relative position between the node and the target node; by preferentially expanding those nodes that are more likely to be close to the target node, unnecessary node expansion operations are reduced, thereby accelerating path planning; or when the search effect is not good, the expansion area is dynamically adjusted.

[0060] Preferably, the priority formula of the extension is:

[0061] P(n)=ω1·g(n)+ω2·h(n)+ω3·φ(n);

[0062] Wherein, P(n) is the expansion priority of node n, which is used to determine the node to be expanded first in the search;

[0063] Among them, g(n) is the path cost of the node, which represents the actual cost from the starting position to node n;

[0064] Among them, h(n) is the heuristic cost of the node, which represents the estimated distance from node n to the target node;

[0065] Among them, ω1, ω2, ω3 are weight parameters, which are used to adjust the impact of path cost, heuristic cost and expansion factor in priority calculation;

[0066] Among them, φ(n) is the expansion factor, which dynamically adjusts the priority according to the distance between the node and the target node. The calculation formula is as follows:

[0067]

[0068] Among them, β is the adjustment parameter, which controls the rate of change of the expansion factor;

[0069] Among them, γ is the expansion threshold.

[0070] Preferably, the formula for adjusting the dynamic expansion area in Step 8 is as follows:

[0071] E(t+1)=E(t)+ΔE;

[0072] Among them, E(t) is the extended area at time t, indicating the scope of the current search;

[0073] Among them, ΔE is the adjustment amount of the expansion area, which is dynamically adjusted according to the search results.

[0074] The present invention has at least the following beneficial effects:

[0075] (1) The present invention provides a wafer extraction path optimization method based on adaptive heuristics and hierarchical search, which reduces the number of motor starts and stops, improves the operating efficiency of automation equipment, extends the life of equipment, and reduces energy consumption while ensuring the shortest path;

[0076] (2) The present invention provides a wafer extraction path optimization method based on adaptive heuristic and hierarchical search, which dynamically adjusts the heuristic value through an adaptive heuristic function, reduces the expansion of invalid paths, accelerates the search process, and thus improves the search efficiency;

[0077] (3) The present invention provides a wafer extraction path optimization method based on adaptive heuristics and hierarchical search, which can divide and simplify the search area layer by layer through a hierarchical path search strategy, thereby reducing the complexity of path planning;

[0078] (4) The present invention provides a wafer extraction path optimization method based on adaptive heuristics and hierarchical search, which adopts an adaptive node expansion mechanism in a dynamic and complex environment to ensure the accuracy and optimality of the path planning results;

[0079] (5) The present invention provides a wafer extraction path optimization method based on adaptive heuristics and hierarchical search. The algorithm can be adjusted according to environmental characteristics and historical path data to adapt to dynamic changes and complex obstacle distribution in different scenarios;

[0080] (6) The present invention provides a wafer extraction path optimization method based on adaptive heuristics and hierarchical search, which can be widely used in complex path planning scenarios such as robot navigation, automated production equipment, and drone path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 The flowchart of the wafer extraction path optimization algorithm based on adaptive heuristics and hierarchical search of the present invention;

[0082] Figure 2 It is the workflow diagram of the adaptive heuristic function of the present invention;

[0083] Figure 3 A first schematic diagram of hierarchical path search according to the present invention;

[0084] Figure 4 A second schematic diagram of hierarchical path search according to the present invention;

[0085] Figure 5 A comparison diagram of the improved algorithm of the present invention and other algorithm paths;

[0086] Figure 6 This is a comparison diagram of the number of nodes of the improved algorithm of the present invention;

[0087] Figure 7 This is a comparison diagram of the path length of the improved algorithm of the present invention;

[0088] Figure 8 This is a time comparison chart of the improved algorithm of the present invention. DETAILED DESCRIPTION

[0089] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0090] Example

[0091] This paper proposes a wafer extraction path optimization method based on adaptive heuristics and hierarchical search, which realizes the global optimal planning of wafer extraction path by dynamically adjusting the search weight and path search level. This algorithm can not only quickly find the shortest path, but also effectively reduce the number of motor starts and stops, and optimize the equipment operation efficiency. Figure 1 As shown, its main technical solution includes the following steps:

[0092] Step 1 (S1): Initialization:

[0093] Set the starting point and end point, define the grid environment and initial parameters;

[0094] Step 2 (S2): Heuristic value calculation:

[0095] Calculate an adaptive heuristic function value based on the current node to guide the search;

[0096] Step 3 (S3): Determine whether the destination has been reached:

[0097] If the current node is the end point, jump to Step 10, stop and output the path, otherwise continue;

[0098] Step 4 (S4): Path cost calculation:

[0099] Calculate the moving cost from the current node to the next node according to the path cost formula;

[0100] Step 5 (S5): Determine node expansion. Determine whether to expand the current node based on the adaptive expansion condition, and give priority to expanding the node that is most likely to be close to the end point;

[0101] Step 6 (S6): Hierarchical search path:

[0102] Gradually perform hierarchical search from the outer layer to the inner layer, and perform independent path planning at each layer to reduce the search scope;

[0103] Step 7 (S7): Path backtracking:

[0104] If the search at a certain level fails or is invalid, return to Step 5, reselect the previous level node, and return to the previous level search;

[0105] Step 8 (S8): Update the search area:

[0106] Dynamically adjust the expansion area, adjust the search parameters according to the results, and proceed to the next step;

[0107] Step 9 (S9): Continue or terminate the search:

[0108] Determine whether the search needs to continue. If there is no valid path, return to Step 4. If the path is valid, output the final path and complete the planning.

[0109] Step 10 (S10): Result output:

[0110] The path planning results are output to guide the equipment to extract the wafer along the optimized path.

[0111] ①. For Step 1, initialize:

[0112] Define the map grid, starting point and end point, and set initial parameters, such as heuristic function coefficients λ1, λ2, etc. This step provides the basic environment for subsequent operations.

[0113] ② For Step 2: Heuristic value calculation:

[0114] The heuristic function is calculated for each node, combined with the local density and historical path information to provide a basis for selecting the optimal node in the next step.

[0115] The calculation formula of the adaptive heuristic function is as follows:

[0116] h′(n)=h(n)+λ1D(n)+λ2H(n);

[0117] Among them, h′(n) is the adaptive heuristic value used to guide the search process;

[0118] Among them, h(n) is the traditional heuristic value, usually using the Euclidean distance, and the calculation formula is as follows:

[0119]

[0120] Among them, x goal and goal is the coordinate of the target node;

[0121] Among them, x n and n is the coordinate of the current node;

[0122] Among them, λ1 and λ2 are adjustment parameters used to control the influence of D(n) and H(n) on the adaptive heuristic value.

[0123] D(n) is the local environment density adjustment item, which reflects the density of obstacles around the current node. The calculation formula is as follows:

[0124]

[0125] Among them, α is the adjustment parameter, which controls the influence of density on the heuristic function;

[0126] Among them, d good (n) is the distance from the current node to the n nearest good chips;

[0127] Among them, d bad (n) is the distance from the current node n to the nearest obstacle (bad chip);

[0128] H(n) is the historical path correction term, which is used to give priority to the valid path direction that has been passed. The calculation formula is as follows:

[0129]

[0130] Among them, P history is the node set in the historical path;

[0131] Among them, dist(n,p) is the distance between the current node n and the historical node p.

[0132] Adaptive heuristic function workflow diagram, such as Figure 2 shown.

[0133] ③. For Step 4-Step 8:

[0134] Find the optimal path from the starting point to the target point, optimize the path according to the motor start and stop conditions, and reduce the start and stop operations of the equipment. At the same time, determine whether the start and stop need to be optimized during path expansion. If the number of starts and stops is too many, readjust the path.

[0135] 1. Hierarchical path search

[0136] like Figure 3 and Figure 4 As shown in the figure, hierarchical path search is to divide the entire search area into several layers, and each layer is searched independently according to its complexity. The specific steps include: first search the outer layer, and then gradually move to the inner layer. The search of each layer can reduce the search space, thereby speeding up the overall process of path planning.

[0137] Through hierarchical search, the scope of each search is reduced and path planning is accelerated.

[0138] The calculation formula of the hierarchical path cost function is as follows:

[0139]

[0140] Among them, C total is the total path cost, which represents the total cost from the starting position to the target node;

[0141] Among them, C layer (S i ) is the path cost of the i-th layer, which consists of the actual cost and the heuristic cost.

[0142] C layer (S i ) is calculated as follows:

[0143] C layer (S i )=g(S i )+h(S i );

[0144] Among them, g(S i ) is from the starting position to layer S iThe cumulative cost represents the actual cost on the path;

[0145] Among them, h(S i ) is a heuristic estimate within the layer, usually using Manhattan distance or Euclidean distance.

[0146] g(S i ) is calculated as follows:

[0147]

[0148] Among them, cost(n,n prev ) is the node n and the previous node n prev The actual cost between.

[0149] 2. Adaptive node expansion

[0150] The adaptive node expansion mechanism dynamically adjusts the priority of nodes to be expanded according to the relative position between the node and the target node. By preferentially expanding those nodes that are more likely to be close to the target node, unnecessary node expansion operations are reduced, thereby speeding up path planning. Or when the search effect is not good, the expansion area is dynamically adjusted.

[0151] The priority formula for the extension is:

[0152] P(n)=ω1·g(n)+ω2·h(n)+ω3·φ(n);

[0153] Wherein, P(n) is the expansion priority of node n, which is used to determine the node to be expanded first in the search;

[0154] Among them, g(n) is the path cost of the node, which represents the actual cost from the starting position to node n;

[0155] Among them, h(n) is the heuristic cost of the node, which represents the estimated distance from node n to the target node;

[0156] Among them, ω1, ω2, ω3 are weight parameters, which are used to adjust the impact of path cost, heuristic cost and expansion factor in priority calculation;

[0157] Among them, φ(n) is the expansion factor, which dynamically adjusts the priority according to the distance between the node and the target node. The calculation formula is as follows:

[0158]

[0159] Among them, β is the adjustment parameter, which controls the rate of change of the expansion factor;

[0160] Among them, γ is the expansion threshold, which determines which nodes are expanded first.

[0161] The dynamic expansion area adjustment formula is as follows:

[0162] E(t+1)=E(t)+ΔE;

[0163] Among them, E(t) is the extended area at time t, indicating the scope of the current search;

[0164] Among them, ΔE is the adjustment amount of the expansion area, which is dynamically adjusted according to the search results.

[0165] In the adaptive heuristic function, we introduce a dynamic weight adjustment mechanism to respond to environmental changes more flexibly. At the same time, adding dynamic feedback of the local environment can improve the real-time performance of the algorithm. By introducing dynamic cost adjustment of each layer, the heuristic estimate is fine-tuned according to the search results of each layer, making it more adaptable in complex environments. The priority evaluation of node expansion is optimized, and the expansion strategy is made more adaptable by introducing information that comprehensively considers the node history and environmental status.

[0166] The improved algorithm of the present invention is compared in terms of the number of mobile nodes, path length, and algorithm time. The results are as follows:

[0167] The present invention compares the improved algorithm with other algorithm paths, such as Figure 5 shown. Figure 5 In the figure, the upper left is the starting point, the lower right is the end point, and the lines from top to bottom represent Bellman-Ford, Greedy Best-First, AHA*, DFS, Dijkstra, BFS, Floyd-Warshall, and A*.

[0168] ①, Comparison of the number of nodes of the improved algorithm, such as Figure 6 As shown in Table 1:

[0169] Table 1 Comparison of the number of nodes of the improved algorithm

[0170] Algorithm Name Number of mobile nodes A* 120 Dijkstra 150 Bellman-Ford 170 DFS 180 BFS 160 Floyd-Warshall 140 Greedy Best-First 190 AHA* 100

[0171] As can be seen from Table 1, the improved algorithm performs very well in terms of the number of mobile nodes, with only 100 mobile nodes, which is much lower than other algorithms. This shows that it can intelligently select the nodes to be searched during path planning, avoiding unnecessary node expansion. In contrast, the number of mobile nodes of the A* and Floyd-Warshall algorithms are 120 and 140 respectively, which are not bad, but still cannot catch up with the improved algorithm proposed in the present invention. Some other algorithms, such as Dijkstra and Bellman-Ford, are more loose in node expansion, resulting in an increase in the number of mobile nodes, which is not only time-consuming but also consumes more resources. Therefore, the advantage of the improved algorithm in reducing the number of mobile nodes not only makes path search more efficient, but also reduces the number of motor starts and stops, saving time and energy for the task execution of automated machines.

[0172] ②Comparison of path lengths of the improved algorithm, such as Figure 7 As shown in Table 2:

[0173] Table 2 Comparison of moving path lengths of improved algorithms

[0174]

[0175]

[0176] As can be seen from Table 2, the improved algorithm also performs well in terms of the length of the moving path, with a path length of only 45, the shortest among all algorithms. This means that it can find a better path, reduce the overall moving distance, and make the automated machine more efficient when extracting wafers. The path lengths of other algorithms are generally long, especially the Bellman-Ford algorithm, which reaches 70, indicating that its path is redundant in some cases. Although the A* algorithm also performs well, with a path length of 50, it is still slightly longer than the improved algorithm in the present invention. In general, the optimization of the path length of the improved algorithm allows tasks to be completed faster, improves the operating efficiency of automated equipment, extends equipment life, and reduces energy consumption. It also provides a good direction for improvement in system design.

[0177] ③、Comparison of time of improved algorithm, such as Figure 8 As shown in Table 3:

[0178] Table 3. Time comparison of improved algorithms

[0179] Algorithm Name Tracking time / ms A* 100 Dijkstra 200 Bellman-Ford 300 DFS 250 BFS 280 Floyd-Warshall 220 Greedy Best-First 400 AHA* 90

[0180] As shown in Table 3, in terms of running time, the execution time of the improved algorithm is 90 milliseconds, which is significantly better than other algorithms, reflecting its excellent computational efficiency. In contrast, the running time of the Greedy Best-First algorithm reaches 400 milliseconds, showing its high overhead when processing large amounts of data. This result emphasizes that in applications that require fast response, the improved algorithm can complete tasks more effectively and improve user experience. Overall, the advantage of the improved algorithm in running time provides an excellent choice for system design with high real-time requirements.

[0181] The present invention can adjust the search strategy more intelligently through the innovative methods of adaptive heuristic functions, hierarchical path search and adaptive node expansion, and improve the efficiency of overall path planning by using real-time environmental feedback and historical experience. This dynamic adjustment mechanism will ensure that the algorithm has better adaptability and flexibility in complex and changing environments, thereby effectively shortening the search time and improving the performance of path planning. In the process of wafer automation extraction, it can significantly reduce the path length, reduce the number of motor starts and stops, improve the operating efficiency of the equipment, reduce energy consumption, and extend the service life of the equipment.

[0182] Based on the above points, the present invention can be applied to:

[0183] 1. Autonomous driving navigation system:

[0184] Combined with adaptive heuristic functions, the search path can be adjusted in real time to dynamically respond to traffic data and obstacle changes, providing more efficient navigation.

[0185] 2. Robot autonomous cruising:

[0186] Through the hierarchical path search method, the robot can perform path optimization at different levels, such as combining the global path with local adjustments, to improve its adaptability in dynamic environments.

[0187] 3. Intelligent logistics scheduling:

[0188] Hierarchical path search can effectively handle path planning problems in large logistics networks, optimizing layer by layer from global scheduling to local path adjustment to achieve more efficient cargo flow.

[0189] 4. Drone patrol and monitoring:

[0190] Adaptive node expansion can give priority to areas with higher flight safety, and quickly respond to changes in the flight environment by dynamically adjusting the expansion area.

[0191] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention.

[0192] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A wafer extraction path optimization method based on adaptive heuristics and hierarchical search, characterized in that: The following steps are involved: Step 1, initialization: Set the starting point and end point, define the grid environment and initial parameters; Step 2, heuristic value calculation: Calculate an adaptive heuristic function value based on the current node to guide the search; Step 3, determine whether it has reached the end point: If the current node is the end point, jump to Step 10, stop and output the path, otherwise continue; Step 4, path cost calculation: Calculate the moving cost from the current node to the next node according to the path cost formula; Step 5, determine the node expansion: Determine whether to expand the current node based on the adaptive expansion condition, and give priority to expanding the node that is most likely to be close to the end point; Step 6, hierarchical search path: Gradually conduct hierarchical search from the outer layer to the inner layer, and perform independent path planning at each layer to reduce the search scope; Step 7, path backtracking: If the search at a certain level fails or is invalid, return to Step 5, reselect the previous level node, and return to the previous level search; Step 8, update the search area: Dynamically adjust the expansion area, adjust the search parameters according to the results, and proceed to the next step; Step 9, continue or terminate the search: Determine whether the search needs to continue. If there is no valid path, return to Step 4. If the path is valid, output the final path and complete the planning. Step 10, result output: The path planning results are output to guide the equipment to extract the wafer along the optimized path.

2. According to claim 1, a wafer extraction path optimization method based on adaptive heuristics and hierarchical search is characterized in that: The calculation formula of the adaptive heuristic function in Step 2 is as follows: h′(n)=h(n)+λ1D(n)+λ2H(n); Among them, h′(n) is the adaptive heuristic value used to guide the search process; Where h(n) is the traditional heuristic value.

3. The wafer extraction path optimization method based on adaptive heuristics and hierarchical search according to claim 2 is characterized in that: The h(n) is usually calculated using the Euclidean distance, and the calculation formula is as follows: Among them, x goal and goal is the coordinate of the target node; Among them, x n and n is the coordinate of the current node; Among them, λ1 and λ2 are adjustment parameters used to control the influence of D(n) and H(n) on the adaptive heuristic value; D(n) is the local environment density adjustment item, which reflects the density of obstacles around the current node; H(n) is the historical path correction term, which is used to give priority to valid path directions that have been used.

4. The wafer extraction path optimization method based on adaptive heuristic and hierarchical search according to claim 3 is characterized in that: The calculation formula of D(n) is as follows: Among them, α is the adjustment parameter, which controls the influence of density on the heuristic function; Among them, d good (n) is the distance from the current node to the n nearest good chips; Among them, d bad (n) is the distance from the current node n to the nearest obstacle; The calculation formula of H(n) is as follows: Among them, P history is the node set in the historical path; Among them, dist(n,p) is the distance between the current node n and the historical node p.

5. The wafer extraction path optimization method based on adaptive heuristic and hierarchical search according to claim 1 is characterized in that: The hierarchical search path divides the entire search area into several layers, and each layer is searched independently according to its complexity. The specific steps are as follows: First, search the outer layer and then gradually move to the inner layer. The search of each layer can reduce the search space. Through layered search, the scope of each search can be reduced and path planning can be accelerated. The calculation formula of the hierarchical path cost function is as follows: Among them, C total is the total path cost, which represents the total cost from the starting position to the target node; Among them, C layer (S i ) is the path cost of the i-th layer, which consists of the actual cost and the heuristic cost.

6. The wafer extraction path optimization method based on adaptive heuristics and hierarchical search according to claim 5 is characterized in that: The C layer (S i ) is calculated as follows: C layer (S i )=g(S i )+h(S i ); Among them, g(S i ) is from the starting position to layer S i The cumulative cost represents the actual cost on the path; Among them, h(S i ) is a heuristic estimate within the layer, usually using Manhattan distance or Euclidean distance.

7. The wafer extraction path optimization method based on adaptive heuristic and hierarchical search according to claim 1 is characterized in that: The g(S i ) is calculated as follows: Among them, cost(n,n prev ) is the node n and the previous node n prev The actual cost between.

8. The wafer extraction path optimization method based on adaptive heuristic and hierarchical search according to claim 1 is characterized in that: The mechanism of adaptive node expansion is to dynamically adjust the priority of the node to be expanded according to the relative position between the node and the target node; reduce unnecessary node expansion operations by preferentially expanding those nodes that are more likely to be close to the target node, thereby accelerating path planning; or dynamically adjust the expansion area when the search effect is not good.

9. The wafer extraction path optimization method based on adaptive heuristics and hierarchical search according to claim 8 is characterized in that: The priority formula of the extension is: P(n)=ω1·g(n)+ω2·h(n)+ω3·φ(n); Wherein, P(n) is the expansion priority of node n, which is used to determine the node to be expanded first in the search; Among them, g(n) is the path cost of the node, which represents the actual cost from the starting position to node n; Among them, h(n) is the heuristic cost of the node, which represents the estimated distance from node n to the target node; Among them, ω1, ω2, ω3 are weight parameters, which are used to adjust the impact of path cost, heuristic cost and expansion factor in priority calculation; Among them, φ(n) is the expansion factor, which dynamically adjusts the priority according to the distance between the node and the target node. The calculation formula is as follows: Among them, β is the adjustment parameter, which controls the rate of change of the expansion factor; Among them, γ is the expansion threshold.

10. The wafer extraction path optimization method based on adaptive heuristic and hierarchical search according to claim 1, characterized in that: The formula for adjusting the dynamic expansion area in Step 8 is as follows: E(t+1)=E(t)+ΔE; Among them, E(t) is the extended area at time t, indicating the scope of the current search; Among them, ΔE is the adjustment amount of the expansion area, which is dynamically adjusted according to the search results.

Images