Distributed Unmanned Vehicle Task Allocation via Auction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing networks of unmanned vehicles for intelligence, surveillance, and reconnaissance missions are limited by centralized decision-making configurations, which are not scalable, suffer from data integrity and availability issues, and result in inflexible mission management.
Innovation Solution
A distributed architecture for unmanned vehicles that enables autonomous operation and cooperative task allocation through an automatic decision level with a task allocation module, task refining module, and task manager module, allowing vehicles to participate in auctions for task allocation and manage tasks autonomously, reducing the need for centralized supervision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If centralized decision-making configuration is used, then mission management control is improved, but scalability and flexibility deteriorate
Solution Approach 1:
The centralized decision-making architecture is segmented into distributed decision-making units across multiple unmanned vehicles. Each vehicle is equipped with autonomous decision-making capabilities, dividing the central control function into independent distributed nodes that can operate autonomously while maintaining coordination through communication protocols.
Solution Approach 2:
The control architecture is inverted from centralized-topdown to distributed-bottom-up. Instead of a central controller directing all vehicles, individual vehicles make autonomous decisions and coordinate with peers through market-based negotiation mechanisms, reversing the traditional command hierarchy.
2Ease of operation
If centralized decision-making configuration is used, then mission management control is improved, but system scalability deteriorates
Solution Approach 1:
The monolithic centralized control system is segmented into modular autonomous decision-making units that can be independently added or removed from the network. Each unit follows standardized communication protocols, allowing seamless integration of new vehicles without increasing overall system complexity.
Solution Approach 2:
The autonomous decision-making architecture provides universal functionality across vehicles of different types and capabilities. The market-based negotiation mechanism and standardized interfaces allow heterogeneous vehicles to participate equally in task allocation, making the system scalable without requiring specialized control logic for each vehicle type.
3Ease of operation
If centralized decision-making configuration is used, then task allocation control is improved, but data integrity and availability deteriorate
Solution Approach 1:
The centralized task allocation function is segmented into distributed autonomous decision-making capabilities across multiple vehicles. Each vehicle independently evaluates tasks and negotiates allocation through market-based mechanisms, eliminating single points of failure and ensuring task allocation continuity even when individual vehicles or communication links fail.
Solution Approach 2:
The system implements continuous feedback loops where vehicles report task execution status, sensor data, and operational state to the network. This distributed feedback mechanism maintains data integrity through multiple independent reporting channels and enables real-time reconfiguration of task allocations based on changing conditions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention provides a co-operative network of unmanned vehicles that participate in co-operative task allocation. Each unmanned vehicle comprises a computer system with an executive level configured to control motion of the unmanned vehicle and an automatic decision level configured to communicate with other unmanned vehicles, to receive task descriptions, to participate in task allocation, and to provide descriptions of tasks to be performed to the executive level for execution. Each unmanned vehicle submits a bid for each task it can perform, the bid reflecting an execution cost for the task. The bids are compared and the task effectively auctioned to the unmanned vehicle with a lowest execution cost. Each unmanned vehicle builds a task plan from its allocated tasks, decomposes the associated task descriptions into elementary task descriptions and forwards the elementary task descriptions in a format suitable for the executive level to execute.