CN-122022284-A - Task distribution system for multi-unmanned forklift collaborative operation

CN122022284ACN 122022284 ACN122022284 ACN 122022284ACN-122022284-A

Abstract

The invention relates to the technical field of unmanned vehicle cooperative control, and discloses a task distribution system for cooperative operation of a plurality of unmanned forklifts. The system comprises a task receiving and analyzing module, a forklift state sensing and fusing module, a dynamic load balancing decision module and a task allocation and execution coordination module. The invention fundamentally changes the limitation that the traditional task allocation only pays attention to a single distance or time index by introducing a multi-objective optimization mechanism integrating energy consumption prediction and load balance degree. The system can dynamically sense the real-time electric quantity and the energy consumption state of each forklift, and actively seek the optimal balance point between the whole energy consumption of the system and the balance of the workload of each forklift when tasks are distributed, so that the phenomenon that part of the forklifts consume early electric quantity due to continuous high-load work and other forklifts are idle in resource waste is effectively avoided, and the whole continuous operation capacity and time of a motorcade are remarkably improved.

Inventors

CHEN ZHEN
TANG BO
WEI SHAOPENG
MA HAIWEN
ZHANG WEI
LI JINYUAN
ZHOU PENG
CHEN NING
LI YU

Assignees

中铁城建集团华东建设有限公司

Dates

Publication Date: 20260512
Application Date: 20260114

Claims (10)

1. The utility model provides a task distribution system of many unmanned fork truck collaborative jobs which characterized in that includes: the task receiving and analyzing module is used for receiving a carrying task instruction from the upper management system, analyzing the task instruction, and extracting task attribute information, wherein the task attribute information comprises a task starting position, a target position, cargo weight, task emergency degree and expected completion time limit; the forklift state sensing and fusing module is used for collecting multidimensional state data of each unmanned forklift in the motorcade in real time, wherein the multidimensional state data comprises real-time position coordinates, advancing speed, battery residual capacity percentage, energy consumption rate per unit time, current load and health state identification of the forklift; the dynamic load balancing decision module is used for executing multi-objective optimization calculation integrating energy consumption prediction and load balancing degree based on task attribute information and real-time multi-dimensional state data of all unmanned forklifts so as to generate an optimal forklift assignment scheme aiming at a current task to be distributed; The task allocation and execution coordination module is used for receiving the optimal forklift assignment scheme, issuing a corresponding task instruction to the assigned unmanned forklift, monitoring the task execution state and triggering a task redistribution flow when abnormality occurs in the task execution process.
2. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 1, wherein the dynamic load balancing decision module comprises an energy consumption prediction unit, a load balancing degree evaluation unit and a multi-objective optimization solving unit; The energy consumption prediction unit is used for predicting the total energy consumption value of the whole task executing process of each candidate unmanned forklift and the current task to be allocated; the load balance degree evaluation unit is used for evaluating the load balance condition of the whole motorcade in a period of time in the future after the current task to be distributed is distributed to the specific candidate unmanned forklift; The multi-objective optimization solving unit is used for taking the minimum task execution total energy consumption and the maximum system load balance degree as two optimization targets, establishing a multi-objective optimization mathematical model, solving the mathematical model by adopting a non-dominant ordering genetic algorithm with elite strategy, generating a group of pareto optimal solutions in the algorithm iteration process, and selecting a final solution from the pareto optimal solutions set as an optimal forklift assignment scheme according to a preset decision rule.
3. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 2, wherein the prediction process of the energy consumption prediction unit comprises: According to the task starting position, the target position and the candidate forklift real-time position, and by combining preset map topology information, an optimal travelling path is planned; And calculating a theoretical total energy consumption value for completing the task based on the length of the planned path, the current load and the weight of the goods to be carried of the candidate forklift, gradient change factors on the path and an energy consumption model of the forklift.
4. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 2, wherein the evaluation process of the load balancing evaluation unit comprises: constructing an equilibrium function taking the estimated residual capacity of each forklift as a core index, wherein the function quantifies the discrete degree of the estimated residual capacity of each forklift relative to the average estimated residual capacity of a motorcade after task allocation; the lower the degree of dispersion, the higher the characterization load balancing.
5. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 2, wherein the decision rule is set to prioritize a solution that contributes most to load balancing, and when there are multiple solutions that are comparable in load balancing, a solution is selected in which the total energy consumption for executing the task is less.
6. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 1, wherein the forklift state sensing and fusion module is further provided with a state data validity checking mechanism; the state data validity checking mechanism carries out outlier detection and smooth filtering processing on the collected original state data, and carries out data fusion on the real-time position and speed information of the forklift by utilizing a Kalman filtering algorithm so as to improve the accuracy and reliability of the state data.
7. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 1, wherein the task allocation and execution coordination module monitors a task execution state, the task execution state specifically monitored including a task progress percentage, a deviation between actual energy consumption of the forklift and a predicted value, and a battery power reduction rate; When the fact that the actual energy consumption is continuously higher than the predicted value and reaches a preset threshold value, or the battery electric quantity reduction rate is abnormally accelerated, or a fault signal actively reported by a forklift is received, the task is judged to be abnormal in execution, and then a task reassignment flow is triggered.
8. The task allocation system for collaborative operations with multiple unmanned forklifts according to claim 7, wherein the task allocation process resubmits the current abnormal task to the dynamic load balancing decision module and marks the task urgency thereof as the highest level, and simultaneously temporarily moves the forklifts executing the abnormality out of the available fleet list until the status thereof returns to normal.
9. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 1, further comprising a historical data learning and model updating module; The historical data learning and model updating module continuously records historical task allocation records, actual energy consumption data of each forklift and task completion conditions; And (3) based on the accumulated historical data, calibrating and updating the energy consumption model parameters in the energy consumption prediction unit by adopting a multiple linear regression method so as to improve the accuracy of energy consumption prediction.
10. The task allocation system for collaborative operation of a multi-unmanned forklift according to claim 1, wherein the system operates entirely in a hierarchical cyclic decision architecture comprising a strategic configuration layer, a tactical scheduling layer, and an operations execution layer; The strategic configuration layer is responsible for setting optimization target weight of a system level, an expected threshold value of load balancing degree and definition rules of various task priorities; the tactical scheduling layer corresponds to a dynamic load balancing decision module, which responds to new task arrival or a system state significant change event at a time interval of seconds or minutes to execute a round of task allocation decision; the operation execution layer corresponds to the task allocation and execution coordination module and each unmanned forklift body and is responsible for millisecond-level or second-level task instruction receiving, path tracking and state feedback.

Description

Task distribution system for multi-unmanned forklift collaborative operation Technical Field The invention belongs to the technical field of unmanned vehicle cooperative control, and particularly relates to a task distribution system for cooperative operation of a plurality of unmanned forklifts. Background In the fields of industrial automation and intelligent warehouse logistics, the unmanned forklift system is used as core equipment for realizing material handling automation, and a task allocation and cooperative control strategy of the unmanned forklift system has important significance for improving overall operation efficiency and system robustness. The task distribution system for the multi-unmanned forklift collaborative operation aims at dynamically and efficiently distributing the carrying tasks from the upper management system to each forklift unit in the motorcade for execution through a reasonable scheduling mechanism. The prior art generally employs static or simple dynamic allocation algorithms based on shortest distance or optimal task completion time. However, the method has the remarkable limitation that the method only takes the task distance or the path length as a core optimization target, and generally ignores the multi-dimensional dynamic constraints such as the real-time energy consumption state, the battery residual quantity, the load change and the like of a forklift in the task execution process. In the peak period of the task, the system load cannot be distributed in an equalizing manner according to the actual working capacity and the energy level of each forklift, the high-energy forklift can be continuously distributed with heavy tasks so as to accelerate the electricity consumption of the high-energy forklift, and the low-energy forklift can be in an idle or low-efficiency state, so that the overall working efficiency of the system is reduced sharply and the task completion time is prolonged. Therefore, how to design a task allocation method capable of comprehensively considering task demands and multi-dimensional real-time states of forklifts, realizing dynamic balance of system loads and finally guaranteeing long-term efficient and stable operation of a multi-unmanned forklift system under a complex warehouse operation environment has become a key technical problem to be solved by a person skilled in the art. Disclosure of Invention The invention aims to solve the technical problems that in the prior art, a multi-unmanned forklift task distribution system only takes distance or time as an optimization target, and dynamic constraints such as real-time energy consumption, battery electric quantity and load change of a forklift are ignored, so that the defects of uneven system load and reduced overall efficiency are overcome. The invention provides a task distribution system for collaborative operation of a multi-unmanned forklift, which aims to realize comprehensive consideration of task demands and multidimensional real-time states of the forklift, achieve dynamic balance of system loads and ensure long-term efficient and stable operation of the multi-unmanned forklift. The technical scheme of the invention is that the task distribution system for the multi-unmanned forklift collaborative operation comprises a task receiving and analyzing module, a forklift state sensing and fusing module, a dynamic load balancing decision module and a task distribution and execution coordination module. The task receiving and analyzing module is used for receiving the carrying task instruction from the upper management system, analyzing the task instruction, and extracting task attribute information, wherein the task attribute information comprises a task starting position, a target position, cargo weight, task emergency degree and expected completion time limit. The forklift state sensing and fusing module is used for collecting multidimensional state data of each unmanned forklift in the motorcade in real time, wherein the multidimensional state data comprises, but is not limited to, real-time position coordinates of the forklift, advancing speed, battery residual capacity percentage, energy consumption rate per unit time, current load and health state identification. The dynamic load balancing decision module is used for executing multi-objective optimization calculation integrating energy consumption prediction and load balancing degree based on task attribute information and real-time multi-dimensional state data of all unmanned forklifts so as to generate an optimal forklift assignment scheme aiming at a current task to be distributed. The task allocation and execution coordination module is used for receiving an optimal forklift assignment scheme, issuing a corresponding task instruction to an assigned unmanned forklift, monitoring the task execution state and triggering a task redistribution flow when an abnormality occurs in the task execution process. Further, the dynamic load bal