A comprehensive planning method for omnidirectional AGV path and posture

Abstract

A method for comprehensive planning of the path and posture of an omnidirectional AGV, characterized in that it comprises the following steps: Step 1: constructing an indoor map model using a topological map method; Step 2: using a depth-first search algorithm to perform path planning under the body posture constraint conditions of each road section; Step 3: calculating the body posture adjustment site of the AGV car in the planned path under the body posture constraint conditions of the omnidirectional AGV; Step 4: using each body posture adjustment site as a segmentation point, dividing the entire path into several sections, and finding the intersection of the body posture sets of each basic section constituting each route; Step 5: sequentially controlling the omnidirectional AGV car to pass through each road section with a planned body posture, and adjusting the body posture at the corresponding node until the car arrives at the end. The present invention can avoid the situation in which the car cannot pass smoothly due to the body posture restriction during driving, and help the car adjust the body posture and arrive at the end smoothly.

Description

Technical Field

[0001] The present invention relates to a path and posture comprehensive planning method, specifically a path and posture comprehensive planning method for an omnidirectional AGV, which aims to solve the problem of body posture adjustment of the omnidirectional AGV during driving. Background Art

[0002] In recent years, with the development and progress of intelligent technology, Automated Guided Vehicle (AGV) as a means of transportation has been widely used in the catering industry and is one of the important means to realize smart restaurants.

[0003] As a type of AGV, the advantage of the omnidirectional AGV lies in its flexibility. It can support the vehicle to move forward, backward, left, right, and spin on the spot freely according to actual work needs during driving. It can solve some problems that ordinary wheel food delivery carts may encounter during work.

[0004] For example, when the food delivery cart is serving food, it needs to be parked next to the table in the most convenient position for the customer, so that the user can easily pick up the food on the cart. Suppose the customer needs to take food from both sides of the cart. If the food delivery cart uses ordinary wheels, the cart can only enter the station with the front of the car. Figure 1 As shown in the right picture, the front of the car is facing the dining table, and the car body is still some distance away from the dining table. Customers need to get off their seats to get the dishes, which is not only inconvenient for customers, but also may affect the customer experience. Figure 1 The posture shown in the left picture is to drive sideways towards the dining table. Assuming that the dishes are evenly placed on the food delivery cart, the customers sitting at the table do not even need to leave their seats and can easily take the dishes on the food delivery cart.

[0005] Another example Figure 2 As shown, Figure 2 This is a simple restaurant map model designed by this method. Assume that the car needs to deliver food from the food pickup area to table 1, that is, drive from station 1 to station 25. If the car is currently at station 12 and detects that the route 12-18 is blocked, the food delivery car will wait at station 12 for the road to be clear before continuing to drive. However, if the rear road 7-12 happens to be congested at this time, and there are obstacles approaching from the front, and the front and rear roads are blocked, then the food delivery car with ordinary wheels may be unable to move, thus falling into a deadlock and requiring manual handling by on-site staff. If the food delivery car is an omnidirectional AGV, then the car can move sideways or spin in place to enter station 11 or station 13 to avoid obstacles.

[0006] Although the omnidirectional AGV brings a lot of convenience during use and can freely change the body posture according to needs, due to the body posture restrictions of some stations or sections during the driving process, the car is only allowed to pass in a certain posture. For example, a narrow passage only allows the car to pass with the narrowest side of the body parallel to the road, which brings many inconveniences to the driving of the omnidirectional AGV.

[0007] To this end, the present invention proposes a comprehensive path and posture planning method for an omnidirectional AGV, which can plan the posture adjustment nodes of the AGV during driving according to the restrictions on the body posture of the vehicle at stations or sections during driving while planning the path, thereby helping the vehicle to adjust its posture and arrive at the destination smoothly. Summary of the invention

[0008] The purpose of the present invention is to provide a path and posture comprehensive planning method for an omnidirectional AGV in response to the body posture adjustment problem encountered during the driving process. The method can plan the posture adjustment nodes of the AGV during the driving process according to the body posture restriction conditions that may be encountered during the driving process, thereby helping the vehicle to adjust its posture and arrive at the destination smoothly.

[0009] The technical solution of the present invention is as follows:

[0010] A path and posture comprehensive planning method for an omnidirectional AGV comprises the following steps:

[0011] Step 1: Use the topological map method to build an indoor map model. The topological map method is to represent some important areas in the environment with nodes, and connect the nodes with line segments to represent the path. In addition to representing the geographical location of the area, the node also contains restrictions such as the AGV's entry posture and the maximum rotation angle.

[0012] Step 1.1: Construct a map site model. Divide a number of sites on all paths in the restaurant that allow cars to pass through, and these sites can be used to divide the path into several basic sections. These site models must at least contain the following information: site number, site location information, neighboring site numbers, the angle at which the car is allowed to rotate, and the site's restrictions on the AGV's body posture when entering the station. The basis for dividing sites in this method is generally the fork or turning point of the road and the starting and ending points of the section where the road conditions change.

[0013] Step 1.2: Construct a road map model and divide all drivable roads into several basic sections to describe the location information of the restaurant road, the vehicle posture restriction information when the vehicle passes through the road, the narrowness of the road, etc. Assume that the channel width in this method is Width road , the narrow channel generally refers to the situation described in formula (1):

[0014] Width road <Width car +2*Width safe (1)

[0015] Width car Refers to the widest distance of the vehicle body perpendicular to the driving direction during the vehicle's driving process. safe Refers to the safe distance between the car and the road boundary, which is set to 0.5m in the present invention. The reason why narrow passages are considered is that it is assumed that there is a 1.2m wide road, and the car is 1m long and 0.5m wide. In this case, if the car chooses to move sideways, that is, the body moves perpendicular to the road direction, then the car is only 10cm away from the road boundary, which will affect the safe passage of the car. Therefore, it is necessary to indicate the narrow passage to limit the body posture.

[0016] Step 2: Use the depth-first search algorithm to plan the path under the vehicle posture constraints of each road section. Since this method uses the topological map method to construct the map model, the sites in the topological map can be regarded as the vertices of the graph in graph theory, and the lines between the sites, that is, the routes, can be regarded as the edges in the graph. The depth-first search algorithm, as one of the commonly used algorithms in graph theory, can find the path between two nodes that meets the requirements, and then calculate the sum of the distances of the edges that make up the path. The distance of each path can be obtained, which meets the needs of this method. The vehicle posture constraints described in the present invention generally refer to the vehicle body direction allowed for the car to pass through a certain road section.

[0017] Step 2.1: Calculate the distance matrix. If there are n sites in the map, then define an n·n two-dimensional matrix e in advance. e[i,j] represents the distance between site i and site j. If site i can be reached directly without passing through any other sites, then e[i,j] is equal to the distance between site i and site j. If site i cannot be reached directly or i is equal to j, then e[i,j] can be equal to a very large fixed value.

[0018] Step 2.2: Input the starting station and the target station. Starting from the starting station, traverse the unvisited stations along the current station. Before visiting, first determine whether the current station allows the adjustment of the vehicle body posture. If the current station allows the adjustment of the vehicle body posture, it means that no matter what the previous vehicle body posture is, the car can be adjusted at the current station, so that the car can reach the station to be visited from the current station; if the vehicle body posture cannot be adjusted, the intersection of the vehicle body posture set allowed by the previous section and the vehicle body posture set of the section to be visited is calculated. If the intersection is empty, it means that the car cannot reach the station and abandons the station. If the intersection is not empty, it means that the car can reach the station and visit the station. When there are no unvisited stations, return to the previous station and continue to try to visit other stations until all stations have been visited. During the visit, if the target station is visited, all the nodes passed will be recorded and stored in the backup path, and the sum of all distances will be recorded. After the visit, the path with the shortest distance is selected as the current path, and the remaining paths are stored in the backup queue in order according to the distance. When the current path is unreachable, the paths are switched from the backup queue in turn.

[0019] Step 3: Under the relevant constraints of the road, calculate the body posture adjustment points of the car in the planned path. The body posture described in this method only considers straight-moving and sideways, and does not consider the oblique movement. The straight-moving mentioned in this method means that the body direction is parallel to the road direction, such as Figure 1 As shown in the left figure, the lateral displacement means that the direction of the vehicle body is perpendicular to the direction of the road. Figure 1 As shown in the picture on the right.

[0020] Step 3.1: First, combine the restaurant map model with the planned path, traverse each station along the way, find the narrow passages in the path, and calculate the restriction conditions for the food delivery AGV to pass through each basic section. The restriction conditions include: the body posture allowed for the food delivery AGV to pass through the section, the posture of the AGV entering the station, and the maximum angle that the AGV can rotate at the station.

[0021] Step 3.2: Then calculate the body posture adjustment points of the car in the planned path.

[0022] First, let's make an assumption: how far can the car travel without changing its body posture? Figure 2 As shown, assuming that it is necessary to drive from station 2 to the charging area to charge, the planned path is 2-3-4-26. Starting from station 2, through step 3.1, the set P of body postures that allow the car to pass through each basic section can be obtained. end (end indicates the identifier of the end station of the basic road segment), the vehicle posture set P of route segment 2-3 3 The vehicle posture set P of route segment 3-4 4Find the intersection, the result is S. If S is not empty, it means that the car can drive from station 2 to station 4 without changing its body posture. At this time, add S to the set P of route segments 4-26. 26 Find the intersection. If the result is empty, it means that the car needs to change its body posture at least once when driving from station 2 to station 26. If station 4 allows the car to adjust its body posture, then the station that allows the car to adjust its body posture is at station 4. If the body posture adjustment is not allowed, it is necessary to reversely traverse each station from station 4 to station 2 in the planned path. The first station that allows the body posture adjustment is the adjustment station that the car needs on the current section of the road. Then use this body posture adjustment station as the current new starting point and repeat the above process until all stations are traversed. You can get all the body posture adjustment stations. If all stations cannot adjust the body posture, it means that the path is unreachable, jump out of step 3 and switch routes. The body posture set described in this embodiment is, Figure 2 As shown, taking the actual direction corresponding to the top of the map as the reference, set it to v. Assume that the car needs to drive from station 7 to station 24 at this time. This is a road from bottom to top in the map. Assume that this section of road allows the car to move longitudinally and laterally, then the set of body postures allowed for this section of road in the map is {v,u,-u,-v}, where v means that the front of the car is allowed to point upward in the map, which means going straight forward in this section of the road; u means that the front of the car is allowed to point to the right in the map, which means moving sideways in this section of the road; -u means that the front of the car is allowed to point to the left in the map, which means moving sideways in this section of the road; -v means that the front of the car is allowed to point downward in the map, which means reversing in this section of the road. Among them, v,u,-u,-v do not require the front of the car to always maintain this angle, but allow the car to adjust its direction based on this direction without deviating from the road.

[0023] Step 4: Divide the road sections according to the vehicle body posture adjustment site and calculate the vehicle body posture of each section.

[0024] Each vehicle posture adjustment station is used as a segmentation point to divide the entire path into several sections, each of which is recorded as Road i (i is the serial number of each route); each route Road i It is composed of several basic sections. i The intersection of the allowed vehicle posture sets for each basic road section is obtained to obtain the set Status i , Status i It is to allow small cars to pass through the route Road i If the allowed body posture is not unique, you can get it from Status i Select a body posture that best suits the current direction of travel. If Status i for It indicates that the current path is unreachable, and the route is switched. If there is no other route to switch to, step 4 is jumped out and the method ends.

[0025] Step 5: Control the omnidirectional AGV car in turn to pass through each section with the planned body posture, and adjust the body posture at the corresponding node until the car reaches the end point.

[0026] The beneficial effects of this method are:

[0027] The present invention can plan the body posture adjustment nodes of the car according to the body posture restriction conditions that may be encountered during driving, so as to avoid the situation where the car cannot pass smoothly due to the body posture restriction during driving, thereby helping the car to adjust the body posture and arrive at the destination smoothly. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Schematic diagram of the vehicle body posture of the present invention.

[0029] Figure 2 Schematic diagram of a restaurant model according to an embodiment of the present invention.

[0030] Figure 3 It is a flow chart of the present invention.

[0031] Figure 4 It is a distance relationship diagram between the various sites of the present invention.

[0032] Figure 5 It is a flow chart of calculating and adjusting the vehicle body posture nodes and the running posture of each road section of the present invention. DETAILED DESCRIPTION

[0033] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0034] like Figure 1-5 shown.

[0035] The present invention aims at the problem of body posture adjustment encountered by an omnidirectional AGV during driving, and provides a path and posture comprehensive planning method for an omnidirectional AGV. The method can plan the posture adjustment nodes of the AGV during driving in advance according to the body posture restriction conditions that may be encountered during driving, thereby helping the vehicle to adjust its posture and arrive at the destination smoothly.

[0036] Figure 3 It is a flow chart of the present invention, and the specific implementation process is as follows:

[0037] Step 1: Use the topological map method to build an indoor map model. The topological map method is to represent some important areas in the environment with nodes, and connect the nodes with line segments to represent the path. In addition to representing the geographical location of the area, the node also contains restrictions such as the AGV's entry posture and maximum rotation angle.

[0038] Step 1.1 Build a map site model. Divide a number of sites on all paths in the restaurant that allow cars to pass through, and these sites can be used to divide the path into several basic sections. These site models must at least contain the following information: site identification, site location, adjacent site number, the angle at which the car is allowed to rotate, the restrictions on the body posture when entering the station, etc. The basis for setting the site in this method is generally the fork or turning point of the road and the starting and ending points of the section where the road conditions change.

[0039] Step 1.2 builds a road map model, dividing all passable roads into several basic sections to describe the location information of the restaurant road, the body posture restriction information of the vehicle passing through the road, the narrowness of the road, etc. Assume that the channel width in this method is Width road , the narrow channel generally refers to the situation that meets formula (1):

[0040] Width car Refers to the widest distance of the vehicle body perpendicular to the driving direction during the vehicle's driving process. safe Refers to the safe distance between the car and the road boundary. In this embodiment, the safe distance is set to 0.5m.

[0041] Step 2: Use the depth-first search algorithm to perform path planning under the vehicle posture constraints of each road section. Since this embodiment uses the topological map method to construct the map model, the sites in the topological map can be regarded as the vertices of the graph in graph theory, and the lines between the sites, that is, the routes, can be regarded as the edges in the graph. The depth-first search algorithm, as one of the commonly used algorithms in graph theory, can find a path that meets the requirements between two nodes, and then calculate the sum of the distances of the edges that make up the path. The distance of each path can be obtained, which meets the needs of this embodiment. The vehicle posture constraints described in this embodiment generally refer to the vehicle body direction allowed for the car to pass through a certain road section.

[0042] Step 2.1: Calculate the distance matrix. If there are n sites in the map, then define an n·n two-dimensional matrix e in advance. e[i,j] represents the distance between site i and site j. If site i can be reached directly without passing through any other sites, then e[i,j] is equal to the distance between site i and site j. If site i cannot be reached directly or i is equal to j, then e[i,j] can be equal to a large fixed value. Figure 2is a 6×6 distance matrix.

[0043] Step 2.2: Input the starting station and the target station. Starting from the starting station, traverse the unvisited stations along the current station. Before visiting, first determine whether the current station allows the adjustment of the vehicle body posture. If the current station allows the adjustment of the vehicle body posture, it means that no matter what the previous vehicle body posture is, the car can be adjusted at the current station, so that the car can reach the station that needs to be visited. If the vehicle body posture cannot be adjusted, the intersection of the vehicle body posture set allowed by the previous section and the vehicle body posture set of the section to be visited is calculated. If the intersection is empty, it means that the car cannot reach it and abandons the station. If the intersection is not empty, it means that the car can reach it and visit the station. When there are no unvisited stations, return to the previous station and continue to try to visit other stations until all stations have been visited. During the visit, if the target station is visited, all the nodes passed will be recorded and stored in the backup path, and the sum of all distances will be recorded. After the visit, the path with the shortest distance is selected as the current path, and the remaining paths are stored in the backup queue in order according to the distance. When the current path is unreachable, the paths are switched from the backup queue in turn.

[0044] Step 3: Under the relevant constraints of the road, calculate the body posture adjustment points of the car in the planned path. The body posture described in this embodiment only considers straight-moving and sideways movement, and does not consider the oblique movement. The straight-moving in this embodiment means that the body direction is parallel to the road direction, such as Figure 1 As shown in the left figure. The lateral displacement means that the direction of the vehicle body is perpendicular to the direction of the road. Figure 1 As shown in the picture on the right.

[0045] Step 3.1: Combine the restaurant map model with the planned path, traverse each station along the way, calculate the restriction conditions for the food delivery AGV to pass through each basic section, and find and record the narrow passages in the path. The restriction conditions include: the allowed body posture of the food delivery AGV when passing through the section, the posture of the AGV entering the station, the maximum angle that the AGV can rotate at the station, etc. Figure 2 As shown, taking section 7-12 as an example, the car drives from station 7 to station 12, then the restriction conditions of this section of the route include the entry constraint conditions and rotation constraints of station 12 in the station model, and whether the section 7-12 is a narrow passage in the road model.

[0046] Step 3.2: Calculate the adjustment points of the vehicle’s body posture in the planned path. The flowchart is as follows: Figure 5 shown.

[0047] Starting from the starting point, traverse each station on the planned path, and find the intersection of the body postures allowed by each section. If the intersection is not empty, it means that the car can continue to drive without changing the body posture; if the intersection is empty, it means that the car needs to adjust the body posture, and then traverse the passing stations in reverse to find the station that allows the body posture to be adjusted. Then repeat the above process with the body posture adjustment station as the new starting point until the traversal reaches the end.

[0048] First, let's make an assumption: how far can the car travel without changing its body posture? Figure 2 As shown, assuming that it is necessary to drive from station 2 to the charging area to charge, the planned path is 2-3-4-26. Starting from station 2, through step 3.1, the set P of body postures that allow the car to pass through each basic section can be obtained. end (end indicates the identifier of the end station of the basic road segment), the vehicle posture set P of route segment 2-3 3 The vehicle posture set P of route segment 3-4 4 Find the intersection, the result is S. If S is not empty, it means that the car can drive from station 2 to station 4 without changing its body posture. At this time, add S to the set P of route segments 4-26. 26 Find the intersection. If the result is empty, it means that the car needs to change its body posture at least once when driving from station 2 to station 26. If station 4 allows the car to adjust its body posture, then the station that allows the car to adjust its body posture is at station 4. If the body posture adjustment is not allowed, it is necessary to reversely traverse each station from station 4 to station 2 in the planned path. The first station that allows the body posture adjustment is the adjustment station for the car on the current road section. Then use this body posture adjustment station as the current new starting point and repeat the above process until all stations are traversed. You can get all the body posture adjustment stations. If all stations cannot adjust the body posture, it means that the path is unreachable, jump out of step 3 and switch routes. The body posture set described in this embodiment is, Figure 2As shown, taking the actual direction corresponding to the top of the map as the reference, set it as v, assuming that the car needs to drive from station 7 to station 24 at this time. This is a road from bottom to top in the map. Assuming that this section of road allows the car to move longitudinally and laterally, then the set of body postures allowed for this section of road in the map is {v,u,-u,-v}, where v means that the front of the car is allowed to point upward in the map, which means going straight forward in this section of the road, u means that the front of the car is allowed to point to the right in the map, which means moving sideways in this section of the road, -u means that the front of the car is allowed to point to the left in the map, which means moving sideways, and -v means that the front of the car is allowed to point downward in the map, which means reversing in this section of the road. Among them, v,u,-u,-v do not require the front of the car to always maintain this angle, but allow the car to adjust its direction based on this direction without deviating from the road.

[0049] Step 4: Divide the road into sections according to the vehicle posture adjustment stations and narrow roads, and calculate the vehicle posture of each section.

[0050] The entire path is divided into several sections using each vehicle posture adjustment station and the starting and ending points of the narrow road as segmentation points. Each section of the route is recorded as Road i (i is the serial number of each route); each route Road i It is composed of several basic sections. i The intersection of the allowed vehicle posture sets for each basic road section is obtained to obtain the set Status i , Status i It is to allow small cars to pass through the route Road i If the allowed body posture is not unique, you can get it from Status i Select a body posture that best suits the current direction of travel. If Status i for If the current path is unreachable, the route is switched. If there is no other route to switch to, step 4 is skipped and the method ends.

[0051] Take the top of the map as u, such as Figure 2 As shown, the car drives from the kitchen to the restaurant. Assuming that Road n The route segment 2-3-4 is included. 2-3 is a bottom-up segment in the map. The car is allowed to move forward, backward, left, and right. The allowed body posture is {v,u,-v,-u}. The route segment 3-4 is also a bottom-up segment in the map. However, there is usually a gate connecting the restaurant and the kitchen, and the road is relatively narrow. Therefore, the car is only allowed to move forward and backward. The allowed body posture is {v,-v}. Then the intersection Status of the two is {v,-v}, which means that the car can pass through Road with the body posture in Status.n .

[0052] Step 5: Control the omnidirectional AGV car in turn to pass through each section with the planned body posture, and adjust the body posture at the corresponding node until the car reaches the end point.

[0053] The present invention does not involve parts that are the same as the prior art or can be implemented by using the prior art.

Claims

1. A comprehensive planning method for the path and posture of an omnidirectional AGV, characterized by: The following steps are involved: Step 1: Use topological mapping method to build indoor map model; The topological map method is to represent some important areas in the environment with nodes, and connect the nodes with lines to represent the paths; in addition to representing the geographical location of the area, the nodes also contain restrictions; Step 2: Use the depth-first search algorithm to plan the path under the vehicle posture constraints of each road section; when using the topological map method to construct the map model, the sites in the topological map are regarded as the vertices of the graph in graph theory, and the lines between the sites, that is, the routes, are regarded as the edges in the graph. The depth-first search algorithm, as one of the commonly used algorithms in graph theory, can find a path that meets the requirements between two nodes, and then calculate the sum of the distances of the edges that make up the path to obtain the distance of each path; after the planning is successful, the shortest path is selected as the current path, and the remaining paths are stored in the backup queue in turn according to the distance. If the current path is determined to be unreachable in the subsequent steps, the paths are switched from the backup queue in turn; Step 3: Under the omnidirectional AGV body posture constraint condition, calculate the body posture adjustment station of the AGV car in the planned path; the body posture constraint condition refers to the body direction allowed for the car to pass through a certain section of road; the body direction refers to the direction of the front of the car, which is based on the actual direction corresponding to the top of the map, set as v, then the front of the car points to the direction corresponding to the top of the map; u represents that the front of the car points to the right of the map; v or u does not mean that the front of the car needs to always face the top or right of the map, but allows the car to adjust the front direction based on v or u without deviating from the road; the front of the car is designated by the manufacturer or the user as a certain part of the car; the body posture only considers two situations: longitudinal displacement and lateral displacement; the longitudinal displacement means that the body direction of the car is parallel to the road direction, and the lateral displacement is perpendicular to the road direction; Step 4: Use each body posture adjustment station as a segmentation point to divide the entire route into several sections, and find the intersection of the body posture sets of each basic section that constitutes each route. This intersection is the final body posture set that allows the car to pass through the route; if the final body posture set is not unique, select an angle from the set that best matches the current driving direction; Step 5: Control the omnidirectional AGV car to pass through each section in the planned body posture, and adjust the body posture at the corresponding node until the car reaches the end point; The restriction conditions include the AGV entry posture and the maximum rotation angle; When using the depth-first search algorithm to plan a path under the vehicle posture constraints of each road section, first, calculate the distance matrix. If there are n stations in the map, then define an n·n two-dimensional matrix e in advance, where e[i,j] represents the distance between station i and station j. If station i to station j can be reached directly without passing through any other stations, then e[i,j] is equal to the distance value between station i and station j. If station i to station j cannot be reached directly or i is equal to j, then e[i,j] can be equal to a very large fixed value. Secondly, input the starting station and the target station, starting from the starting station, traverse the unvisited stations along the current station. Before visiting, first determine whether the current station allows the adjustment of the vehicle posture. If the current station allows the adjustment of the vehicle posture, it means that regardless of the previous vehicle posture Anything can make the car get adjusted at the current station, so that the car can reach the station that needs to be visited from the current station; if the body posture cannot be adjusted, the body posture set allowed by the previous section and the body posture set of the section to be visited are intersected. If the intersection is empty, it means that the car cannot reach the station and abandons the station. If the intersection is not empty, it means that the car can reach the station and visit the station; when there is no unvisited station, return to the previous station and continue to try to visit other stations until all stations have been visited; during the visit, if the target station is visited, all the nodes passed through are recorded and stored in the backup path, and the sum of all distances is recorded; after the visit, the path with the shortest distance is selected as the current path, and the remaining paths are stored in the backup queue according to the distance. When the current path is unreachable, the paths are switched from the backup queue in turn; The use of the topological map method to construct an indoor map model includes: constructing a site map model and constructing a road map model; constructing a site map model is to set up sites at key nodes of the road to describe the location information of the site, the adjacent site numbers, the AGV vehicle entry posture restrictions, and the maximum rotation angle allowed; the key nodes refer to road forks or turning points and the starting and ending points of sections where road conditions change; the construction of the road map model is to divide all passable roads into several basic sections according to the sites to describe the location information of the road and whether the road is narrow; the location information of the road refers to the coordinate information of the starting and ending points of the section, and a narrow channel refers to a channel width that is less than the vehicle width plus twice the safety distance, wherein the vehicle width refers to the length of the narrower side of the vehicle body, and the safety distance refers to the distance between the vehicle and the road boundary; The calculation of the body posture adjustment site of the AGV in the planned path under the omnidirectional AGV body posture constraint condition refers to: first, according to the map model and the planned path, traverse each site passed through, and calculate the restriction conditions for the AGV to pass through each basic section; secondly, use the body posture constraint condition to calculate the posture adjustment site in the planned path; traverse the planned path, and find the intersection of the body postures allowed by each section one by one. If the intersection is not empty, It means that the car can keep driving without changing its body posture; if the intersection is empty, it means that the car needs to adjust its body posture, and then traverse the passed stations in reverse to find the station that allows the body posture to be adjusted; then repeat the above process with the station where the body posture is adjusted as the new starting point until the traversal reaches the end; the body posture set is based on the actual direction corresponding to the top of the map, set to v, assuming that the car needs to drive from one station to another at this time, assuming that this section of the road allows the car to move longitudinally and laterally, then the body posture set allowed for this section of the road in the map is {v,u,-u,-v}, where v means the front of the car is allowed to point upwards in the map, which means going straight forward on this road; u means the front of the car is allowed to point to the right in the map, which means moving sideways on this road; -u means the front of the car is allowed to point to the left in the map, which means moving sideways on this road; -v means the front of the car is allowed to point downwards in the map, which means driving in reverse on this road. v,u,-u,-v does not require the front of the car to always maintain this angle, but allows the car to adjust its direction based on this direction without deviating from the road.

Images