Multi-Robot Mission Assignment Using GA and Terrain-Aware CNN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-robot systems rely heavily on human intuition and experience for mission assignment, leading to inefficiencies and potential errors, especially in rapidly changing environments, where automation technologies like genetic algorithms (GA) and convolutional deep neural networks can improve mission determination but require advanced methods for optimizing device operations.
Innovation Solution
A method using a genetic algorithm (GA) to determine device missions by calculating an object function based on external situation and status information, iteratively generating solutions, and employing a convolutional neural network to assign missions to robots based on attribute information and terrain data, optimizing mission assignment for multiple devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human intuition and experience are used for mission assignment, then the system is simple to operate, but the accuracy and reliability of mission determination deteriorates
Solution Approach 1:
The patent replaces human intuition and experience-based mission assignment with an automated genetic algorithm system. The GA processes external situation information and device status information through computational operations (selection, crossover, mutation) to determine optimal missions, substituting the mechanical human decision-making process with an automated evolutionary computation system that provides higher accuracy and reliability.
Solution Approach 2:
The system enables autonomous mission determination through the genetic algorithm that automatically processes input information and generates mission assignments without human intervention. The GA performs self-evaluation through the object function calculation and iteratively improves mission assignments through generations, allowing the system to serve itself in determining optimal missions based on current situational data.
2Productivity
If genetic algorithm is used for mission determination, then the productivity and accuracy of mission assignment is improved, but the device complexity increases
Solution Approach 1:
The genetic algorithm implementation is segmented into distinct functional modules: external situation information processing, device status information processing, object function calculation, selection operation, crossover operation, and mutation operation. This segmentation allows the complex GA system to be managed through independent, well-defined components that can be processed and optimized separately, improving overall system productivity while managing complexity through modular design.
Solution Approach 2:
The system performs preliminary actions by pre-defining the object function that evaluates mission assignment quality, and pre-establishing the GA parameters (population size, generation limit, crossover rate, mutation rate) before execution. This preliminary configuration enables the GA to efficiently proceed through generations without requiring complex real-time adjustments, thereby improving mission assignment productivity while keeping the operational complexity manageable.
3Adaptability or versatility
If the number of devices in the multi-robot system is increased, then the versatility and capability of the system is improved, but the difficulty of monitoring and control increases
Solution Approach 1:
The patent replaces manual monitoring and control of multiple devices with an automated genetic algorithm system that processes status information from all devices simultaneously. The GA-based mission assignment system automatically evaluates the state of each device and determines appropriate missions without requiring human operators to manually track and coordinate each device, thereby enabling the system to handle increased device counts while reducing monitoring difficulty.
Solution Approach 2:
The genetic algorithm implementation serves multiple functions: it processes external situation information, evaluates device status, calculates object functions for different mission scenarios, performs selection/crossover/mutation operations, and generates mission assignments for multiple devices. This universal approach allows the same system architecture to effectively manage any number of devices, improving system versatility while maintaining manageable control complexity through standardized processing.
Data Source
AI summary
Provided is a method of determining a mission of a device in an electronic device, including identifying external situation information and status information of a plurality of devices, identifying an object function related to a degree of fitness of a device for a mission, for each set of a plurality of solutions of a first generation for the plurality of devices, calculating a value of the object function using the external situation information and the status information, based on the value, determining a set of a plurality of solutions of a next generation for the plurality of devices through a genetic algorithm (GA), and based on a first solution from a set of a plurality of solutions of a second generation by repeating the calculating the value and the determining the set of the plurality of solutions of the next generation, determining a mission for the plurality of devices.