In-field transfer robot intelligent scheduling system and algorithm thereof

Abstract

The invention discloses an in-field transfer robot intelligent scheduling system and an algorithm thereof, and the system comprises a vehicle cloud interface module which is used for receiving the real-time position, state and sensing data of a vehicle, and issuing a cloud scheduling instruction; the task management module is used for receiving, analyzing, decomposing and managing tasks; the scheduling decision module is used for formulating a scheduling scheme; the path planning module is used for generating an optimal transportation path and an obstacle avoidance strategy; the scheduling execution module is used for coordinating multi-vehicle cooperation and task execution; and the safety management module is used for guaranteeing vehicle and system safety. The operation efficiency can be remarkably improved, the waiting time is shortened, and the turnover speed is increased; operation cost is reduced, resource utilization is optimized, and energy consumption is reduced; the safety is improved, and the accident risk is reduced; intelligent scheduling is realized, and manual intervention is reduced; and data support is provided to assist in management decision making.

Description

Technical Field

[0001] This invention relates to the field of logistics vehicle scheduling technology, specifically to an intelligent scheduling system for in-plant handling robots. Background Technology

[0002] In-plant logistics refers to the entire process of storing, transporting, packaging, processing, loading, unloading, handling, and managing goods within a specific area. Currently, in-plant logistics generally suffers from the following problems: low efficiency of manual dispatching, frequent traffic congestion, insufficient equipment utilization, high safety risks, and serious energy waste. These problems lead to poor multi-vehicle coordination, insufficiently intelligent dispatching decisions, and imperfect safety assurance mechanisms. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings and deficiencies of existing technologies by providing an intelligent, efficient, and safe on-site handling robot scheduling system and algorithm.

[0004] To achieve the above objectives, the technical solution adopted by this invention is: an intelligent scheduling system for in-field handling robots, the innovation of which lies in comprising: The vehicle-cloud interface module is used to receive real-time vehicle location data, status data, and various sensing data; and to issue dispatch instructions from the cloud. The task management module is used to receive, analyze, decompose, and manage tasks; the scheduling decision module is used to decide on scheduling schemes. The path planning module is used to generate optimal transportation routes and obstacle avoidance schemes; The scheduling and execution module is used to coordinate multi-vehicle collaborative operations and task execution; The safety management module is used to ensure vehicle safety and system security.

[0005] Furthermore, the vehicle-to-cloud interface module includes: The vehicle and roadside sensing device access management unit adopts high concurrency and large connection technology to support 100,000 vehicles and roadside sensing devices to be online in real time; through parallel expansion, it can also support more vehicles and roadside sensing devices to be online in real time. The data transmission unit is used to transmit real-time vehicle location data, status data, and various sensing data; it is also used to issue dispatch instructions from the cloud. The emergency message transmission unit is used to transmit messages and instructions related to early warning and safety.

[0006] Furthermore, the task management module includes: Task decomposition unit: Breaks down complex logistics tasks into basic operational units; Priority management unit, which sorts tasks according to their urgency and importance.

[0007] Furthermore, the scheduling decision module includes: The resource allocation unit allocates tasks based on vehicle type, load capacity, proximity principle, and load balancing principle; The task coordination unit handles the scheduling logic for multi-vehicle collaborative tasks; The intelligent learning unit can continuously learn and optimize the algorithm based on user intervention in the task allocation scheme.

[0008] Furthermore, the path planning module includes: Static map building unit, used to build digital maps of the site environment; Dynamic route planning unit calculates the optimal transportation route in real time; Obstacle avoidance strategy unit, generates dynamic obstacle avoidance schemes; The traffic control unit manages traffic flow and right-of-way allocation within the venue.

[0009] Furthermore, the scheduling execution module includes: The real-time scheduling unit executes task allocation and path planning results. The status monitoring unit monitors the task execution status in real time; The exception handling unit handles unexpected situations during execution; The performance optimization unit continuously optimizes scheduling strategies and execution efficiency.

[0010] Furthermore, the security management module includes: The vehicle safety unit is responsible for traffic flow control, monitoring of vehicle internal and external safety parameters, vehicle safety early warning, and emergency response to safety accidents. The system security unit is responsible for system security protection and security auditing.

[0011] An intelligent scheduling system algorithm for in-field handling robots, the innovation of which is that it includes the following algorithm flow: S1. Heuristic path algorithm: Its formula is f(n)=g(n)+w(n)*h(n) (1) Where: f(n) is the estimated total cost of reaching the target from the starting point through node n; g(n) is the actual cost from the starting point to node n; h(n) is the heuristic cost estimate from node n to the target; w(n) is a dynamic weighting factor; S2. Integrated Motion Constraints: Consider vehicle kinematic characteristics, including minimum turning radius, acceleration limit, speed limit, vehicle size, road slope, load capacity, road surface slipperiness, wind direction, wind speed, etc. Modified actual cost function: g(n)=g_distance(n)+α* g_kinematic(n) (2) The range of values ​​for α is [0.1, 0.5]. g_distance(n)= base_cost * distance + turn_cost * angle_change (3) in: base_cost: Base movement cost coefficient (1.0 or 1.414) distance: distance turn_cost: Turning cost coefficient (0.1~0.3) angle_change: Steering angle g_kinematic(n)=turn_radius_cost+velocity_cost+ acceleration_cost turn_radius_cost=k_r * max(0, (R_min / R_current - 1)) (4) in: k_r: Turning radius coefficient (0.2) R_min: Minimum turning radius R_current: Current turning radius velocity_cost=k_v*|v_current-v_target| / v_max (5) in: k_v: Velocity variation coefficient (0.15) v_max: Maximum allowed speed acceleration_cost=k_a * |a_current| / a_max (6) in: k_a: Acceleration coefficient (0.15) a_max: Maximum allowable acceleration; S3. Introducing a spatiotemporal dimension: obstacle motion prediction, time window division, collision risk assessment, and real-time path adjustment; S4. Strategies to optimize search efficiency: including hierarchical search strategies and improvements to heuristic functions; The hierarchical search strategy includes global path planning and local path optimization; The improved formula for the heuristic function is as follows: h(n)=max(h_manhattan(n), h_euclidean(n))+β*h_obstacle(n) (7) The range of values ​​for β is: [0.1, 0.3] h_manhattan(n)=|x_goal - x_current| + |y_goal - y_current| (8) h_euclidean(n)=sqrt((x_goal - x_current)² + (y_goal - y_current)²) (9) h_obstacle(n)=k_obs*(obstacle_density + obstacle_proximity)(10) in: x_goal: The X-coordinate of the target location y_goal: The Y-coordinate of the target position x_current: The X coordinate of the current position y_current: The Y-coordinate of the current position k_obs: Obstacle influence coefficient (0.2) obstacle_density: Local obstacle density [0,1] obstacle_proximity: Normalized value of the nearest obstacle distance [0,1].

[0012] Furthermore, the following factors are dynamically adjusted for w(n): distance from the target point, density of surrounding obstacles, degree of path congestion, task priority, and user intervention.

[0013] The beneficial effects of this invention are: 1. This invention can improve work efficiency: reduce waiting time and increase turnaround speed; 2. This invention can reduce operating costs by optimizing resource utilization and reducing energy consumption; 3. This invention can improve safety: reduce the risk of accidents and ensure personnel safety; 4. This invention enables intelligent operation: reducing manual intervention and improving the level of automation; 5. This invention can optimize management by providing data support and assisting in decision analysis. Attached Figure Description

[0014] Figure 1 This is a system architecture diagram of the present invention; Figure 2This is one of the overall flowcharts of the present invention; Figure 3 This is the second overall flowchart of the present invention; Figure 4 This is the third overall flowchart of the present invention; Figure 5 This is the fourth overall flowchart of the present invention; Figure 6 This is a flowchart of the heuristic path algorithm of the present invention; Figure 7 This is a flowchart of the intelligent learning algorithm for the scheduling scheme of this invention. Detailed Implementation

[0015] The invention will now be further described with reference to the accompanying drawings.

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Example

[0017] See Figure 1-2 This system consists of six modules that work collaboratively to achieve intelligent scheduling throughout the entire process, from task assignment to execution monitoring. The system achieves data exchange through a vehicle-to-cloud interface and, through modules for task management, scheduling decision-making, route planning, scheduling execution, and safety management, completes efficient, safe, and adaptive on-site logistics scheduling. The main modules include: The vehicle-cloud interface module is used to receive real-time vehicle location data, status data, and various sensing data; and to issue dispatch instructions from the cloud. The task management module is used to receive, analyze, decompose, and manage tasks; the scheduling decision module is used to decide on scheduling schemes. The path planning module is used to generate optimal transportation routes and obstacle avoidance schemes; The scheduling and execution module is used to coordinate multi-vehicle collaborative operations and task execution; The safety management module is used to ensure vehicle safety and system security.

[0018] The vehicle-to-cloud interface module includes: The vehicle and roadside sensing device access management unit adopts high concurrency and large connection technology to support 100,000 vehicles and roadside sensing devices to be online in real time; through parallel expansion, it can also support more vehicles and roadside sensing devices to be online in real time. The data transmission unit is used to transmit real-time vehicle location data, status data, and various sensing data; it is also used to issue dispatch instructions from the cloud. The emergency message transmission unit is used to transmit messages and instructions related to early warning and safety.

[0019] The task management module includes: Task decomposition unit: Breaks down complex logistics tasks into basic operational units; Priority management unit, which sorts tasks according to their urgency and importance.

[0020] The scheduling decision module includes: The resource allocation unit allocates tasks based on vehicle type, load capacity, proximity principle, and load balancing principle; The task coordination unit handles the scheduling logic for multi-vehicle collaborative tasks; The intelligent learning unit can continuously learn and optimize the algorithm based on user intervention in the task allocation scheme.

[0021] The route planning module includes: Static map building unit, used to build digital maps of the site environment; Dynamic route planning unit calculates the optimal transportation route in real time; Obstacle avoidance strategy unit, generates dynamic obstacle avoidance schemes; The traffic control unit manages traffic flow and right-of-way allocation within the venue.

[0022] The scheduling and execution module includes: The real-time scheduling unit executes task allocation and path planning results. The status monitoring unit monitors the task execution status in real time; The exception handling unit handles unexpected situations during execution; The performance optimization unit continuously optimizes scheduling strategies and execution efficiency.

[0023] The security management module includes: The vehicle safety unit is responsible for traffic flow control, monitoring of vehicle internal and external safety parameters, vehicle safety early warning, and emergency response to safety accidents. The system security unit is responsible for system security protection and security auditing.

[0024] This system integrates information resources and arranges paths rationally through multiple sets and modules, thereby improving its level of intelligence and automation. Example

[0025] See Figure 3-4 This algorithm achieves intelligent path planning and scheduling optimization through the following steps: A heuristic search algorithm is used to dynamically adjust weight factors to adapt to the real-time environment; Introducing kinematic constraints into the cost function improves path feasibility; Dynamic obstacle avoidance and real-time path adjustment are achieved through spatiotemporal analysis; By employing hierarchical search and heuristic function improvement strategies, the algorithm efficiency can be enhanced. Specifically as follows: S1. Heuristic path algorithm: Its formula is f(n)=g(n)+w(n)*h(n) (1) Where: f(n) is the estimated total cost of reaching the target from the starting point through node n; g(n) is the actual cost from the starting point to node n; h(n) is the heuristic cost estimate from node n to the target; w(n) is a dynamic weighting factor; S2. Integrated Motion Constraints: Consider vehicle kinematic characteristics, including minimum turning radius, acceleration limit, speed limit, vehicle size, road slope, load capacity, road surface slipperiness, wind direction, wind speed, etc. Modified actual cost function: g(n)=g_distance(n)+α* g_kinematic(n) (2) The range of values ​​for α is [0.1, 0.5]. g_distance(n)= base_cost * distance + turn_cost * angle_change (3) in: base_cost: Base movement cost coefficient (1.0 or 1.414) distance: distance turn_cost: Turning cost coefficient (0.1~0.3) angle_change: Steering angle g_kinematic(n)=turn_radius_cost+velocity_cost+ acceleration_cost turn_radius_cost=k_r * max(0, (R_min / R_current - 1)) (4) in: k_r: Turning radius coefficient (0.2) R_min: Minimum turning radius R_current: Current turning radius velocity_cost=k_v*|v_current-v_target| / v_max (5) in: k_v: Velocity variation coefficient (0.15) v_max: Maximum allowed speed acceleration_cost=k_a * |a_current| / a_max (6) in: k_a: Acceleration coefficient (0.15) a_max: Maximum allowable acceleration; S3. Introducing a spatiotemporal dimension: obstacle motion prediction, time window division, collision risk assessment, and real-time path adjustment; S4. Strategies to optimize search efficiency: including hierarchical search strategies and improvements to heuristic functions; The hierarchical search strategy includes global path planning and local path optimization; The improved formula for the heuristic function is as follows: h(n)=max(h_manhattan(n), h_euclidean(n))+β*h_obstacle(n) (7) The range of values ​​for β is: [0.1, 0.3] h_manhattan(n)=|x_goal - x_current| + |y_goal - y_current| (8) h_euclidean(n)=sqrt((x_goal - x_current)² + (y_goal - y_current)²) (9) h_obstacle(n)=k_obs*(obstacle_density + obstacle_proximity)(10) in: x_goal: The X-coordinate of the target location y_goal: The Y-coordinate of the target position x_current: The X coordinate of the current position y_current: The Y-coordinate of the current position k_obs: Obstacle influence coefficient (0.2) obstacle_density: Local obstacle density [0,1] obstacle_proximity: Normalized value of the nearest obstacle distance [0,1].

[0026] Among them, w(n) is dynamically adjusted by the following factors: distance from the target point, density of surrounding obstacles, degree of path congestion, task priority, and user intervention.

[0027] In summary, this invention optimizes the system and algorithms, rationally integrates resources, employs intelligent scheduling algorithms for multi-vehicle collaborative operations, and ensures operational safety through real-time monitoring. This invention can significantly improve on-site logistics efficiency, reduce operating costs, and enhance safety, demonstrating significant application value.

[0028] The above description is only used to illustrate the technical solution of the present invention and is not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solution of the present invention, as long as they do not depart from the spirit and scope of the technical solution of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent scheduling system for in-field transport robots, characterized in that, include: The vehicle-to-cloud interface module is used to receive real-time vehicle location data, status data, and various sensing data. Used for issuing scheduling instructions from the cloud, etc. The task management module is used to receive, analyze, decompose, and manage tasks; the scheduling decision module is used to decide on scheduling schemes. The path planning module is used to generate optimal transportation routes and obstacle avoidance schemes; The scheduling and execution module is used to coordinate multi-vehicle collaborative operations and task execution; The safety management module is used to ensure vehicle safety and system security.

2. The intelligent scheduling system for in-field handling robots according to claim 1, characterized in that, The vehicle-to-cloud interface module includes: The vehicle and roadside sensing device access management unit adopts high concurrency and large connection technology to support 100,000 vehicles and roadside sensing devices to be online in real time; through parallel expansion, it can also support more vehicles and roadside sensing devices to be online in real time. The data transmission unit is used to transmit real-time vehicle location data, status data, and various sensing data; it is also used to issue dispatch instructions from the cloud. The emergency message transmission unit is used to transmit messages and instructions related to early warning and safety.

3. The intelligent scheduling system for in-field handling robots according to claim 1, characterized in that, The task management module includes: Task decomposition unit: Breaks down complex logistics tasks into basic operational units; Priority management unit, which sorts tasks according to their urgency and importance.

4. The intelligent scheduling system for in-field handling robots according to claim 1, characterized in that, The scheduling decision module includes: The resource allocation unit allocates tasks based on vehicle type, load capacity, proximity principle, and load balancing principle; The task coordination unit handles the scheduling logic for multi-vehicle collaborative tasks; The intelligent learning unit can continuously learn and optimize the algorithm based on user intervention in the task allocation scheme.

5. The intelligent scheduling system for in-field handling robots according to claim 1, characterized in that, The path planning module includes: Static map building unit, used to build digital maps of the site environment; Dynamic route planning unit calculates the optimal transportation route in real time; Obstacle avoidance strategy unit, generates dynamic obstacle avoidance schemes; The traffic control unit manages traffic flow and right-of-way allocation within the venue.

6. The intelligent scheduling system for in-field handling robots according to claim 1, characterized in that, The scheduling execution module includes: The real-time scheduling unit executes task allocation and path planning results. The status monitoring unit monitors the task execution status in real time; The exception handling unit handles unexpected situations during execution; The performance optimization unit continuously optimizes scheduling strategies and execution efficiency.

7. The intelligent scheduling system for in-field handling robots according to claim 1, characterized in that, The security management module includes: The vehicle safety unit is responsible for traffic flow control, monitoring of vehicle internal and external safety parameters, vehicle safety early warning, and emergency response to safety accidents. The system security unit is responsible for system security protection and security auditing.

8. An algorithm for an intelligent scheduling system of in-field handling robots according to any one of claims 1-7, characterized in that, The algorithm flow includes the following: S1. Heuristic path algorithm: Its formula is f(n)=g(n)+w(n)*h(n) (1) Where: f(n) is the estimated total cost of reaching the target from the starting point through node n; g(n) is the actual cost from the starting point to node n; h(n) is the heuristic cost estimate from node n to the target; w(n) is a dynamic weighting factor; S2. Integrated motion constraints: Consider vehicle kinematic characteristics, including minimum turning radius, acceleration limit, speed limit, vehicle size, road slope, load capacity, road surface slipperiness, wind direction, wind speed, etc. Modified actual cost function: g(n)=g_distance(n)+α* g_kinematic(n) (2) The range of values ​​for α is [0.1, 0.5]. g_distance(n)= base_cost * distance + turn_cost * angle_change (3) in: base_cost: Base movement cost coefficient (1.0 or 1.414) distance: distance turn_cost: Turning cost coefficient (0.1~0.3) angle_change: Steering angle g_kinematic(n)=turn_radius_cost+velocity_cost+ acceleration_cost turn_radius_cost=k_r * max(0, (R_min / R_current - 1)) (4) in: k_r: Turning radius coefficient (0.2) R_min: Minimum turning radius R_current: Current turning radius velocity_cost=k_v*|v_current-v_target| / v_max (5) in: k_v: Velocity variation coefficient (0.15) v_max: Maximum allowed speed acceleration_cost=k_a * |a_current| / a_max (6) in: k_a: Acceleration coefficient (0.15) a_max: Maximum allowable acceleration; S3. Introducing a spatiotemporal dimension: obstacle motion prediction, time window division, collision risk assessment, and real-time path adjustment; S4. Strategies to optimize search efficiency: including hierarchical search strategies and improvements to heuristic functions; The hierarchical search strategy includes global path planning and local path optimization; The improved formula for the heuristic function is as follows: h(n)=max(h_manhattan(n), h_euclidean(n))+β*h_obstacle(n) (7) The range of values ​​for β is: [0.1, 0.3] h_manhattan(n)=|x_goal - x_current| + |y_goal - y_current| (8) h_euclidean(n)=sqrt((x_goal - x_current)² + (y_goal - y_current)²) (9) h_obstacle(n)=k_obs*(obstacle_density + obstacle_proximity)(10) in: x_goal: The X-coordinate of the target location y_goal: The Y-coordinate of the target position x_current: The X coordinate of the current position y_current: The Y-coordinate of the current position k_obs: Obstacle influence coefficient (0.2) obstacle_density: Local obstacle density [0,1] obstacle_proximity: Normalized value of the nearest obstacle distance [0,1].

9. The algorithm for the intelligent scheduling system of in-field handling robots according to claim 8, characterized in that, The following factors are dynamically adjusted for w(n): distance from the target point, density of surrounding obstacles, degree of path congestion, task priority, and user intervention.

Images