Vehicle loading method of roll-on-roll-off ship based on heuristic algorithm
Through the method based on heuristic algorithm, the problems of low scheduling efficiency and low space utilization during the loading process of Ro-Ro-Ro-Shipping Vehicles in car shipping are solved, and safe and smooth vehicle trajectory is generated, loading efficiency and safety are improved, and the loading efficiency and safety are reduced. Fuel consumption and docking time.
Patent Information
- Application Number
- CN202411159607.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-12-06
AI Technical Summary
In the car shipping scenario, there are problems such as low scheduling efficiency and under-utilization of space during the vehicle loading of Roll-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-Ro-R The safety and smoothness of vehicle kinematic constraints and trajectory are ignored.
Using a heuristic algorithm-based method, by collecting environmental data of ship decks and vehicles, establishing geometric space models and kinematic models, designing path planning and trajectory optimization algorithms, and using CBS collision detection strategy and D*Lite path planning algorithm to generate real-time Collision-free vehicle trajectory and virtual simulation is performed through Unity simulation software to optimize the loading process.
It significantly improves the space utilization and loading efficiency of vehicle loading, provides a safe and reliable path, reduces fuel consumption, shortens the vehicle's docking time in the port, and improves the large-scale loading capacity and transportation safety.
Smart Images

Figure CN119089675A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle transportation, and in particular to a method for loading vehicles on a roll-on / roll-off ship based on a heuristic algorithm. Background Art
[0002] As the main means of transportation for people, efficient, safe and low-cost automobile logistics has become a competitive advantage in cross-border vehicle trade. In the international logistics of automobiles, seaborne roll-on / roll-off transportation is the main logistics mode, but roll-on / roll-off transportation has problems such as slow loading speed and low space utilization. Therefore, dense loading and fast loading and unloading have great research significance and application value.
[0003] In the automobile shipping scenario, in view of the problems existing in the vehicle loading process of RoRo ships, such as lack of overall planning, low scheduling efficiency, and underutilized space, it is necessary to carry out scheduling with the goal of minimizing scheduling time and layout space, so as to reduce the time and space costs of loading RoRo ships.
[0004] The vehicle loading problem is decomposed into layout, path finding and scheduling problems. The mainstream solutions are divided into two categories: one is the feasible domain search optimization solution, which uses an exact solution algorithm, and the other is the heuristic positioning optimization solution, which takes the positioning algorithm as the core and the sequencing algorithm based on fitness as the framework. The positioning algorithm is mainly based on geometric space methods such as the leftmost and bottom algorithm, the lowest horizontal line algorithm, and the critical polygon algorithm. The above algorithms can provide a simple and efficient arrangement rule and can quickly construct a high-quality initial solution, but they cannot find the global optimal solution in the face of large-scale and multi-constrained arrangement problems; the sequencing algorithm is mainly based on intelligent optimization algorithms such as genetic algorithms, ant colony algorithms, and simulated annealing algorithms, which have excellent global search and local development capabilities, but the encoding method is unreasonable, the arrangement rules are unclear, and they are greatly affected by parameters.
[0005] The above solutions consider fewer constraints when modeling, the evaluation indicators are not comprehensive, the connection with the scheduling link is not close enough, there is little research on the comprehensive arrangement of tasks, and the improvement strategies adopted often sacrifice the rapidity of the algorithm. The path finding and scheduling solutions undertaken often ignore the kinematic constraints of the vehicle and the safety and smoothness of the trajectory, and lack multi-vehicle scheduling solutions based on single vehicle path planning. Summary of the invention
[0006] According to the technical problems mentioned in the above background technology, a method for loading vehicles on a ro-ro ship based on a heuristic algorithm is provided. The invention can solve the problems of security and timeliness faced by scheduling problems.
[0007] The technical means adopted by the present invention are as follows:
[0008] A method for loading vehicles on a ro-ro ship based on a heuristic algorithm comprises the following steps:
[0009] Step 1: Collect the environmental data of the ship deck and the size data of the vehicle; according to the collected data, expand the vehicle size, establish the geometric space model of the deck and the vehicle, and establish the kinematic model of the vehicle, and propose the objective function of path planning;
[0010] Step 2: Construct the positioning algorithm and sequencing algorithm for vehicle loading to generate the parking points and the order of transportation;
[0011] Step 3: Design a path planning algorithm based on the parking points and transportation order generated in step 2, and design a trajectory generation and trajectory optimization algorithm; traverse the fleet participating in the dispatch through the CBS collision detection strategy, detect collision conflicts in real time, and repeat the CBS collision detection strategy if a collision conflict occurs. For the detected risky vehicles, repeat step 3 and re-plan until the conflict is eliminated;
[0012] Step 4: Import the hull file into the Unity simulation software. When the complete dimension data of the vehicle and cabin are known, the above data is input into the solution algorithm to generate a real-time collision-free vehicle trajectory, and the center of gravity constraint is applied to the heuristic objective function to maintain the center of gravity balance, and a density constraint is applied to the heuristic function.
[0013] Furthermore, in step 1, establishing the geometric space model of the deck and the vehicle includes the following steps:
[0014] Step 11: With the stern of the ship as the vertex, the direction from the ship to the bow and stern is called the longitudinal direction, represented by X1, and the direction from the port to the starboard is called the transverse direction, represented by Y1, and the deck's arrangable space is established to establish a rectangular coordinate system; set the deck's arrangable space to P, the length to L, and the width to M;
[0015] Step 12: Create a geometric model of the vehicle's envelope rectangle; set the vehicle's expansion space to p i , length is l i , width is w i , i = 1, 2, 3…; take the horizontal direction as the X2 axis and the vertical direction as the Y2 axis to establish a rectangular coordinate system, and specify the coordinate of the lower left corner of the vehicle as (x i1 ,y i1 ), the coordinate of the upper right corner of the vehicle is (x j2 ,y j2 );
[0016] Step 13: Establish constraints and objective functions. The constraints are: vehicles are placed without overlap, vehicles can only be placed within the row space, the number and model of vehicles are limited and uniquely determined, the weight of the vehicle on board is ensured to be lower than the ship's load, whether the vehicle is rotated 90 degrees is considered, the unutilized rate of the deck space is minimized, and the instability of the ship is minimized based on the load balance principle on the left and right sides of the ship;
[0017] Step 14: Establish a corresponding kinematic model for the vehicle, and use the grid method to establish a deck model in the cabin. Determine the appropriate scale, i.e., the grid size, based on the vehicle size and the environmental scale, and determine the mathematical model of path planning to record the vehicle information and the time of arrival at the node.
[0018] Further, in step 2, constructing the positioning algorithm and sequencing algorithm for vehicle loading includes the following steps:
[0019] Step 21: The vehicle to be loaded and the obstacle are envelopingly processed by using a critical polygon method, and simplified into graphics such as an enveloping rectangle or an enveloping polygon;
[0020] Step 22: Execute the comparison function of the positioning algorithm to prevent the vehicle from crossing the boundary. If the expected arrangement height of the vehicle exceeds the specified range, repeat step 22 and continue to arrange the next vehicle;
[0021] Step 23: Execute the matching function of the positioning algorithm to measure the compatibility of the vehicle to be sampled and the remaining space;
[0022] Step 24: The overlap judgment of the polygons identified by the NFP critical polygon algorithm is simplified to the judgment of whether the point is within the NFP, and point matching is directly performed during positioning;
[0023] Step 25: Use the destruction operator to remove several vehicles from the current solution, and then use the repair operator to reinsert the removed vehicles into the destroyed solution.
[0024] Furthermore, in step 3, the path generation optimization includes the following steps:
[0025] Step 31: Represent the environment map through a grid map, and complete the path planning according to the environment map and the path planning algorithm;
[0026] Step 32: Run the D*Lite path planning algorithm and perform path correction through the D*Lite path search algorithm; introduce the boundary search operation of the JPS jump point search to reduce the large number of iterations involved in the algorithm search;
[0027] Step 33: Generate a continuous curve for the discrete nodes of the path generated in step 32, select multiple high-order polynomials, consider the smoothness and continuity of the planning generation, select a number of equidistant nodes between each waypoint to impose restrictions, and realize the generation control of the entire path;
[0028] Step 34: Run the IBC trajectory generation algorithm to improve the smoothness of the trajectory based on the fourth-order derivative of the position on the minimum Snap trajectory and the processing of the restriction nodes;
[0029] Step 35: Run the CBS collision detection strategy to traverse the vehicles involved in the dispatch to avoid vehicle collisions and improve the safety of the dispatch;
[0030] Step 36: Run the multi-vehicle parallel scheduling scheme, and use the optimal arrangement sequence solved by the balanced weight arrangement algorithm as the loading order for scheduling. The vehicle moves to the designated point legally according to the planned route, and the loading is completed;
[0031] Further, in step 4, the virtual simulation includes the following steps:
[0032] Step 41: Build the Unity scene;
[0033] Step 42: Place the algorithm under high loading pressure, and the loading effect of the ro-ro vehicle in the Unity scene;
[0034] Step 43: Simulate the driving route solved by the algorithm to achieve a comprehensive evaluation of the algorithm performance.
[0035] Furthermore, in step 4, the hull is constructed by using Rhino and 3D max modeling software.
[0036] Furthermore, the Unity scene includes the hull and internal facilities of the ro-ro ship.
[0037] Compared with the prior art, the present invention has the following advantages:
[0038] The present invention constructs a simulation environment for vehicle loading by collecting deck environment data of real ships and size data of vehicles. A positioning sequencing algorithm for generating vehicle parking points and vehicle driving order is constructed. A path search algorithm, a trajectory generation optimization algorithm, and a collision elimination strategy for vehicle driving trajectory generation, trajectory optimization, and collision conflict detection are constructed. In the method, parking points are generated according to deck and vehicle data, and a driving route is generated according to the vehicle starting point and parking point. The present invention solves practical problems such as slow loading speed, low loading space utilization, and poor loading safety of ro-ro ships, significantly improves the problems of low vehicle loading space utilization and slow algorithm solution speed, and provides a continuous, smooth, short time-consuming, safe and reliable path to achieve the purpose of reducing fuel consumption, improve the efficiency and safety of transportation in vehicle transoceanic trade ro-ro ship vehicle loading scenarios, improve large-volume loading capacity, and shorten the berthing time of vehicle ro-ro ships at ports. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0040] Figure 1 This is the framework diagram of this method;
[0041] Figure 2 This is a deck dimension diagram in an embodiment of the present invention;
[0042] Figure 3 This is a diagram of vehicle size information in an embodiment of the present invention;
[0043] Figure 4 A schematic diagram of vehicle kinematics in an embodiment of the present invention;
[0044] Figure 5Schematic diagram of the matching function in the embodiment of the present invention; wherein, (a) is c=0, u=0, in this case, the vehicle is completely mismatched with the waiting space, the height matching degree c is recorded as 0, the width matching degree η is recorded as 0, and the matching degree u is equal to the sum of the height matching degree and the width matching degree, recorded as 0; (b) is c=1, u=1+η, in this case, the vehicle is still mismatched with the waiting space, the vehicle height is greater than the height of the vehicles on both sides, the height matching degree c is recorded as 1, and the width matching degree η is recorded as the ratio of the width of the waiting space to the width of the vehicle (the vehicle width needs to be less than The width of the waiting space), the matching degree u is equal to the sum of the height matching degree and the width matching degree, recorded as 1+η; (c) is c=1, u=1+η, the vehicle and the waiting space are still not matched, the vehicle height is less than the height of the vehicle on one side, but greater than the height of the vehicle on the other side, the height matching degree c is recorded as 1, the width matching degree η is recorded as the ratio of the width of the waiting space to the width of the vehicle (the width of the vehicle needs to be less than the width of the waiting space), the matching degree u is equal to the sum of the height matching degree and the width matching degree, recorded as 1+η; (d) is c=2, u=2+η, the vehicle and the waiting space are still not matched, the vehicle height is less than the height of the vehicle on one side, but greater than the height of the vehicle on the other side, the height matching degree c is recorded as 1, the width matching degree η is recorded as the ratio of the width of the waiting space to the width of the vehicle (the width of the vehicle needs to be less than the width of the waiting space), the matching degree u is equal to the sum of the height matching degree and the width matching degree, recorded as 1+η; Incomplete matching: the height of the vehicle is approximately equal to the height of the vehicle on the left, but greater than the height of the vehicle on the right. The height matching degree c is recorded as 2, the width matching degree η is recorded as the ratio of the width of the waiting space to the width of the vehicle (the width of the vehicle needs to be smaller than the width of the waiting space), and the matching degree u is equal to the sum of the height matching degree and the width matching degree, recorded as 2+η; (e) is c=2, u=2+η, the vehicle and the waiting space are not completely matched, the height of the vehicle is approximately equal to the height of the right vehicle, but greater than the height of the left vehicle, the height matching degree c is recorded as 2, and the width matching degree η is recorded as the width of the waiting space. The matching degree u is equal to the sum of the height matching degree and the width matching degree, which is recorded as 2+η; (f) is c=3, u=3+η, the vehicle is relatively matched with the waiting space, the vehicle height is approximately equal to the height of the left vehicle, and approximately equal to the height of the right vehicle, the height matching degree c is recorded as 2, the width matching degree η is recorded as the ratio of the waiting space width to the vehicle width (the vehicle width needs to be smaller than the waiting space width), the matching degree u is equal to the sum of the height matching degree and the width matching degree, which is recorded as 3+η;
[0045] Figure 6 Schematic diagram of vehicles to be arranged in an embodiment of the present invention; wherein P1, P2, P3, and P4 are all vehicle rectangles;
[0046] Figure 7 Schematic diagram of the NFP critical polygon algorithm for solving positioning points in an embodiment of the present invention; wherein P4 is a vehicle rectangle;
[0047] Figure 8 Schematic diagram of successful positioning by the NFP critical polygon algorithm in an embodiment of the present invention; wherein P1, P2, P3, and P4 are all vehicle rectangles;
[0048] Fig. 9Schematic diagram of the cavity problem in the embodiment of the present invention; wherein P1, P2, and P3 are all vehicle rectangles;
[0049] Fig.10 Schematic diagram of the NFP critical polygon algorithm for solving the cavity problem in an embodiment of the present invention;
[0050] Fig.11 Schematic diagram of the cavity successfully located by the NFP critical polygon algorithm in an embodiment of the present invention; wherein P1, P2, and P3 are all vehicle rectangles;
[0051] Fig.12 The pseudo code of the NFP positioning algorithm in the embodiment of the present invention;
[0052] Fig.13 The pseudo code of the BA sequencing algorithm in the embodiment of the present invention;
[0053] Fig.14 Schematic diagram of the D*Lite pathfinding algorithm in condition 1 where no safety detection is performed in an embodiment of the present invention; wherein P1, P2, P3, and P4 are waypoints;
[0054] Fig.15 Schematic diagram of the D*Lite pathfinding algorithm for safety detection condition 1 in an embodiment of the present invention; wherein P1, P2, P3, P4, P5, P6, and P7 are waypoints;
[0055] Fig.16 Schematic diagram of the D*Lite pathfinding algorithm in the embodiment of the present invention without performing safety detection condition 2; wherein P1 and P2 are waypoints;
[0056] Fig.17 Schematic diagram of the D*Lite pathfinding algorithm for safety detection condition 2 in an embodiment of the present invention; wherein P1, P*, and P2 are waypoints;
[0057] Fig.18 Schematic diagram of the D*Lite pathfinding algorithm without temporary obstacle marking in an embodiment of the present invention; wherein S and G are waypoints;
[0058] Fig.19 Schematic diagram of the D*Lite pathfinding algorithm for temporary obstacle marking in an embodiment of the present invention; wherein S and G are waypoints;
[0059] Fig. 20 Schematic diagram of the boundary search operation of the D*Lite pathfinding algorithm in an embodiment of the present invention; wherein S, G, P1, P2, P3, P4, and P5 are waypoints;
[0060] Fig.21 The pseudo code of the D*Lite pathfinding algorithm in the embodiment of the present invention;
[0061] Fig. 22 This is a flow chart of a conflict elimination algorithm of a conflict elimination search strategy CBS in an embodiment of the present invention;
[0062] Fig.23 A simplified diagram of a Unity map in an embodiment of the present invention;
[0063] Fig.24 This is a diagram showing the simulation result of the vehicle arrangement using Unity in an embodiment of the present invention;
[0064] Fig.25 It is a schematic diagram of low loading pressure in Matlab simulation environment in an embodiment of the present invention;
[0065] Fig.26 It is a schematic diagram of the middle loading pressure in the Matlab simulation environment in an embodiment of the present invention;
[0066] Fig. 27 It is a schematic diagram of high loading pressure in Matlab simulation environment in an embodiment of the present invention;
[0067] Fig.28 It is a vehicle trajectory diagram in the Unity simulation environment in an embodiment of the present invention. DETAILED DESCRIPTION
[0068] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.
[0069] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0070] like Figure 1 As shown, the embodiment of the present invention discloses a method for loading vehicles on a ro-ro ship based on a heuristic algorithm, comprising the following steps:
[0071] Step 1: Collect the environmental data of the ship deck and the size data of the vehicle. Based on the collected data, expand the vehicle size, establish the geometric space model of the deck and the vehicle, and establish the kinematic model of the vehicle, and propose the objective function of path planning;
[0072] Step 2: Construct the positioning algorithm and sequencing algorithm for vehicle loading to generate the parking points and the order of transportation;
[0073] Step 3: Design a path planning algorithm based on the parking points and transportation order generated in step 2, and design a trajectory generation and trajectory optimization algorithm; traverse the fleet participating in the dispatch through the CBS collision detection strategy, detect collision conflicts in real time, and repeat the CBS collision detection strategy if a collision conflict occurs. For the detected risky vehicles, repeat step 3 and re-plan until the conflict is eliminated;
[0074] Step 4: Build the hull using Rhino and 3D Max modeling software, import the hull file into Unity simulation software, and input the above data into the solution algorithm to generate real-time collision-free vehicle trajectory when the complete size data of the vehicle and cabin are known. The center of gravity constraint is applied to the heuristic objective function to maintain the center of gravity balance, and a density constraint is applied to the heuristic function.
[0075] As a preferred embodiment of the present invention, in step 1, the process of collecting data and establishing a model includes the following steps:
[0076] Step 11: With the stern of the ship as the vertex, the direction from the ship to the bow and stern is called the longitudinal direction, represented by X1, and the direction from the port to the starboard is called the transverse direction, represented by Y1, to establish the deck's layout space and establish a rectangular coordinate system; set the deck's layout space to P, the length to L, and the width to W; Figure 2 As shown, a rectangular coordinate system is established for the layout space of the deck;
[0077] Step 12: Create a geometric model of the vehicle's envelope rectangle; set the vehicle's expansion space to p i , length is l i , width is w i , i = 1, 2, 3…; take the horizontal direction as the X2 axis and the vertical direction as the Y2 axis to establish a rectangular coordinate system, and specify the coordinate of the lower left corner of the vehicle as (x i1 ,y i1 ), the coordinate of the upper right corner of the vehicle is (x j2 ,y j2 );like Figure 3 As shown, a geometric model is established for the envelope rectangle of the vehicle;
[0078] Step 13: Establish constraints and objective functions, requiring that vehicles are placed without overlap, constraining vehicles to be placed only within the row space, ensuring that the number and model of vehicles are limited and uniquely determined, ensuring that the weight of the vehicle after boarding is lower than the ship's load, considering whether the vehicle is rotated 90 degrees, minimizing the unutilized rate of the deck space, and minimizing the instability of the ship based on the load balance principle on the left and right sides of the ship;
[0079] Step 14: Figure 4 As shown, a corresponding kinematic model is established for the vehicle, and the grid method is used to establish a deck model in the cabin. The appropriate scale (grid size) can be determined according to the vehicle size and the environmental scale, and the mathematical model of path planning can be determined to record the vehicle information and the time of arrival at the node, which is convenient for scheduling.
[0080] As a preferred embodiment of the present invention, in step 2, executing the sequencing algorithm and the positioning algorithm includes the following steps:
[0081] Step 21: Simplify the loaded vehicle and obstacles such as columns and steps on the deck into graphics such as enveloping rectangles or enveloping polygons;
[0082] Step 22: Execute the comparison function of the positioning algorithm to prevent the vehicle from crossing the boundary. If the expected arrangement height of the vehicle exceeds the specified range, repeat step 22 and continue to arrange the next vehicle.
[0083] Step 23: Execute the matching function of the positioning algorithm to reduce space waste and measure the compatibility of the vehicles to be sampled with the remaining space; Figure 5 As shown in the figure, the vehicles filled with shaded spaces are the vehicles to be sorted, and the vehicles without shaded spaces are the vehicles that have been sorted. The vehicle in (a) cannot fit into the gap; the vehicles in (b) and (c) can fit in, but are not flush with the heights of the two sides; the vehicles in (d) and (e) are only flush with the height of one side; the vehicle in (f) is consistent with the heights of both sides. The matching degree u is the sum of the width and height matching degrees, η is the ratio of the two, and the height matching degree is defined as c, which measures the compatibility of the vehicles to be sorted and the remaining space, and can be an integer between 0 and 3.
[0084] Step 24: Figure 6-Figure 8 As shown in Figure 2, the NFP critical polygon algorithm simplifies the overlap judgment of the recognition polygon to the judgment of whether the point is within the NFP, and directly performs point matching during positioning. In particular, Figure 9-11 As shown in Figure 2, the NFP critical polygon algorithm can effectively identify the remaining space in the cavity. Fig.12 Shown is the pseudo code of the NFP positioning algorithm.
[0085] Step 25: The sequencing algorithm has transformed the parameter definition and update optimization method, optimized the algorithm's compatibility, introduced destruction and repair operations, and enhanced the algorithm's local optimization ability; the positioning algorithm adopts the BA bat algorithm, which imitates bat echolocation and uses the principle of frequency tuning to achieve automatic scaling of the search range. Changing the encoding and decoding methods is helpful for solving the problem. The initial population is constructed using random initialization. The fitness function imposes penalties on individuals that violate the constraints to meet the loading constraints. The local search operation uses the idea of destruction and repair of the large neighborhood algorithm. Merge operation eliminates unqualified individuals; use the destruction operator to remove several vehicles from the current solution, and then use the repair operator to reinsert the removed vehicles into the destroyed solution; such as Fig.13 Shown is the pseudo code of the BA sequencing algorithm.
[0086] As a preferred embodiment of the present invention, in step 3, the path generation optimization comprises the following steps:
[0087] Step 31: In a known environment, the vehicle completes global path planning by building a complete environment map and path planning algorithm. The environment map is described in the form of a grid map. Black grids represent impassable grids, and white grids represent passable grids.
[0088] Step 32: Run the D*Lite path planning algorithm. The D*Lite path search algorithm path correction operation improves the safety of the path. Figure 14-17 As shown in , when a path that passes through a sharp corner of an obstacle grid or a path at the junction of two obstacle grids is detected, the path is changed to a detour. Figure 18-19 As shown, in order to avoid unnecessary entry into the concave obstacle again when the environment changes and to reduce the number of grids for path search, a temporary obstacle grid marking method is used to avoid unnecessary grid searches; Fig. 20 As shown in the figure, the boundary search operation is introduced to improve the search speed. By introducing the path correction operation, the feasibility of the search result is improved. It adopts the search method from the target node to the starting node. The D*Lite path planning algorithm maintains the g(s) and rhs(s) of each grid (the two are the value function and the estimation function respectively). The consistency of the two indicates that the grid is passable. If a change in the passability of the grid is detected, the D*Lite path planning algorithm will quickly replan. The boundary search operation of the JPS (jump point search) jump point search is introduced to reduce the large number of iterations involved in the algorithm search. Fig.21 Shown is the pseudo code of the D*Lite pathfinding algorithm.
[0089] Step 33: Generate a continuous curve for the discrete nodes of the path generated in step 32, select multiple high-order polynomials, consider the smoothness and continuity of the planning generation, select a number of equidistant nodes between each waypoint to impose restrictions, and realize the generation control of the entire path;
[0090] Step 34: Run the IBC (integral backstepping) trajectory generation algorithm, which is based on the minimum Snap (the fourth-order derivative of the position on the trajectory) and the processing of the restricted nodes to improve the smoothness of the trajectory;
[0091] Step 35: Fig.19 As shown in , the CBS collision detection strategy is run to traverse the vehicles involved in the dispatch to avoid vehicle collisions and improve the safety of the dispatch; Fig.21 Shown is the flow chart of CBS collision detection algorithm.
[0092] Step 36: Run the multi-vehicle parallel scheduling scheme, and use the optimal arrangement sequence solved by the balanced weight arrangement algorithm as the loading order for scheduling. The vehicles will start from the hatch continuously, and the three-dimensional time-space model is used to integrate the time-space curves of all vehicles in the multi-vehicle scheduling strategy. After scanning one by one, it is determined whether there is a collision. If there is a collision, the possible collision type (follow-up collision, side collision or frontal collision). The time-space scheduling model assigns time to the movement trajectory of each vehicle, and adopts strategies such as priority adjustment method and re-search method to eliminate vehicle collision conflicts, and alienate the time-space curves that are about to cross, so as to achieve collision-free parallel movement of multiple vehicles and solve a better scheduling scheme.
[0093] As a preferred embodiment of the present invention, in step 4, the virtual simulation includes the following steps:
[0094] Step 41: Fig.13 As shown in the figure, the scene built in Unity includes the complete hull of the RoRo ship and its internal facilities;
[0095] Step 42: Fig.24 As shown in Figure 2, the algorithm is placed under high loading pressure, and the loading effect of ro-ro vehicles in the Unity environment is shown in Figure 2; Figure 25-27 As shown in the figure, the algorithm is placed under three loading pressures: low, medium and high, and the loading effect of ro-ro vehicles in the Matlab environment;
[0096] Step 43: Fig.28 As shown in the figure, the driving route solved by the algorithm is simulated to achieve a comprehensive evaluation of the algorithm performance.
[0097] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways.
[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for loading ro-ro ships based on a heuristic algorithm, characterized in that: The following steps are involved: Step 1: Collect environmental data of the ship deck and dimension data of the vehicle; Based on the collected data, the vehicle size is expanded, the geometric space model of the deck and vehicle is established, and the kinematic model of the vehicle is established, and the objective function of path planning is proposed; Step 2: Construct the positioning algorithm and sequencing algorithm for vehicle loading to generate the parking points and the order of transportation; Step 3: Design a path planning algorithm based on the parking points and transportation order generated in step 2, and design a trajectory generation and trajectory optimization algorithm; The CBS collision detection strategy is used to traverse the fleet participating in the dispatch and detect collision conflicts in real time. If a collision conflict occurs, the CBS collision detection strategy is repeated. For the detected risky vehicles, step 3 is repeated to re-plan until the conflict is eliminated. Step 4: Import the hull file into the Unity simulation software. When the complete dimension data of the vehicle and cabin are known, the above data is input into the solution algorithm to generate a real-time collision-free vehicle trajectory, and the center of gravity constraint is applied to the heuristic objective function to maintain the center of gravity balance, and a density constraint is applied to the heuristic function.
2. A method for loading ro-ro ships with vehicles based on a heuristic algorithm according to claim 1, characterized in that: In step 1, establishing the geometric space model of the deck and the vehicle includes the following steps: Step 11: With the stern of the ship as the vertex, the direction from the ship to the bow and stern is called the longitudinal direction, represented by X1, and the direction from the port to the starboard is called the transverse direction, represented by Y1, and the deck's arrangable space is established to establish a rectangular coordinate system; set the deck's arrangable space to P, the length to L, and the width to W; Step 12: Create a geometric model of the vehicle's envelope rectangle; set the vehicle's expansion space to p i , length is l i , width is w i , i = 1, 2, 3…; take the horizontal direction as the X2 axis and the vertical direction as the Y2 axis to establish a rectangular coordinate system, and specify the coordinate of the lower left corner of the vehicle as (x i1 ,y i1 ), the coordinate of the upper right corner of the vehicle is (x j2 ,y j2 ); Step 13: Establish constraints and objective functions. The constraints are: vehicles are placed without overlap, vehicles can only be placed within the row space, the number and model of vehicles are limited and uniquely determined, the weight of the vehicle on board is ensured to be lower than the ship's load, whether the vehicle is rotated 90 degrees is considered, the unutilized rate of the deck space is minimized, and the instability of the ship is minimized based on the load balance principle on the left and right sides of the ship; Step 14: Establish a corresponding kinematic model for the vehicle, and use the grid method to establish a deck model in the cabin. Determine the appropriate scale, i.e., the grid size, based on the vehicle size and the environmental scale, and determine the mathematical model of path planning to record the vehicle information and the time of arrival at the node.
3. A method for loading ro-ro ships with vehicles based on a heuristic algorithm according to claim 1, characterized in that: In step 2, constructing the positioning algorithm and sequencing algorithm for vehicle loading includes the following steps: Step 21: The vehicle to be loaded and the obstacle are envelopingly processed by using a critical polygon method, and simplified into an enveloping rectangle or an enveloping polygon; Step 22: Execute the comparison function of the positioning algorithm to prevent the vehicle from crossing the boundary. If the expected arrangement height of the vehicle exceeds the specified range, repeat step 22 and continue to arrange the next vehicle; Step 23: Execute the matching function of the positioning algorithm to measure the compatibility between the vehicle to be sampled and the remaining space; Step 24: The overlap judgment of the polygons identified by the NFP critical polygon algorithm is simplified to the judgment of whether the point is within the NFP, and point matching is directly performed during positioning; Step 25: Use the destruction operator to remove several vehicles from the current solution, and then use the repair operator to reinsert the removed vehicles into the destroyed solution.
4. A method for loading ro-ro ships with vehicles based on a heuristic algorithm according to claim 1, characterized in that: In step 3, the path generation optimization includes the following steps: Step 31: Represent the environment map through a grid map, and complete the path planning according to the environment map and the path planning algorithm; Step 32: Run the D*Lite path planning algorithm and perform path correction through the D*Lite path search algorithm; introduce the boundary search operation of the JPS jump point search to reduce the large number of iterations involved in the algorithm search; Step 33: Generate a continuous curve for the discrete nodes of the path generated in step 32, select multiple high-order polynomials, consider the smoothness and continuity of the planning generation, select a number of equidistant nodes between each waypoint to impose restrictions, and realize the generation control of the entire path; Step 34: Run the IBC trajectory generation algorithm to improve the smoothness of the trajectory based on the fourth-order derivative of the position on the minimum Snap trajectory and the processing of the restriction nodes; Step 35: Run the CBS collision detection strategy to traverse the vehicles involved in the dispatch to avoid vehicle collisions and improve the safety of the dispatch; Step 36: Run the multi-vehicle parallel scheduling plan, and use the optimal arrangement sequence solved by the balanced weight arrangement algorithm as the loading order for scheduling. The vehicle moves legally to the designated point according to the planned route, and the loading is completed.
5. A method for loading ro-ro ships with vehicles based on a heuristic algorithm according to claim 1, characterized in that: In step 4, the virtual simulation includes the following steps: Step 41: Build the Unity scene; Step 42: Place the algorithm under high loading pressure, and the loading effect of the ro-ro vehicle in the Unity scene; Step 43: Simulate the driving route solved by the algorithm to achieve a comprehensive evaluation of the algorithm performance.
6. A method for loading vehicles on a ro-ro ship based on a heuristic algorithm according to claim 1, characterized in that: In step 4, the hull is constructed by using Rhino and 3D max modeling software.
7. A method for loading vehicles on a ro-ro ship based on a heuristic algorithm according to claim 5, characterized in that: The Unity scene includes the hull and internal facilities of the ro-ro ship.
Citation Information
Cited By
Evaluation method for driving route in cabin
CN120146351A
Ferry intelligent drop and pull dispatching and vehicle safety monitoring equipment and method
CN120542876A
Method and system for managing and controlling whole process of transportation of port automobiles
CN120996751A
