Autonomous Agent Radar Countermeasures via Reinforcement Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional radar negation techniques are insufficient for counteracting agile radar networks with advanced sensing capabilities, as they rely on preplanned maneuvers that cannot adapt to dynamic changes in the target radar system.
Innovation Solution
A method using a machine learning module to generate an operational model of a radar network, classify it as friendly or non-friendly, and a reinforcement learning module to generate counter-radar maneuvers, implemented by an autonomous agent, allowing for adaptive and dynamic countermeasures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If preplanned negation maneuvers are used, then the operational framework is simple and straightforward, but the system cannot adapt to dynamic changes in agile radar networks
Solution Approach 1:
The patent applies dynamics by transitioning from static preplanned maneuvers to dynamic adaptive maneuvers. The autonomous agent continuously learns and updates its negation strategies in real-time based on radar network behavior, allowing the system to adapt to changing radar waveforms and tactics while maintaining operational effectiveness.
Solution Approach 2:
The autonomous agent implements self-service by independently learning and generating negation maneuvers without human intervention. The reinforcement learning framework enables the agent to autonomously improve its capabilities through continuous interaction with the radar network environment, adapting to new threats automatically.
2Reliability
If autonomous agents implement precise trajectory computation and execution, then counter-radar maneuvers are time-critical and directionally dependent, but the operational complexity increases
Solution Approach 1:
The patent applies feedback through the reinforcement learning framework, where the autonomous agent receives continuous feedback from radar network responses to its maneuvers. This feedback loop enables the agent to refine its trajectory computation and execution precision over time, learning the optimal maneuvers that maximize negation effectiveness while accounting for radar reactions.
3Adaptability or versatility
If conventional phantom track generation or jamming techniques are used, then the countermeasures can be predetermined, but they cannot account for countermeasures implemented in the target radar system
Solution Approach 1:
The patent applies preliminary action by pre-training the autonomous agent with a diverse library of negation maneuvers and radar response patterns before deployment. This pre-training establishes a foundational knowledge base that enables rapid adaptation during actual operations, reducing the learning time required when facing new radar countermeasures.
Data Source
AI summary
Embodiments of the disclosure provide a machine learning system for identifying and countering non-friendly radar networks. Methods of the disclosure include generating, in a machine learning module, an operational model of a radar network within an environment. An autonomous agent within the environment detects the radar network. The method also includes classifying the radar network as friendly or non-friendly based on the operational model. The method also includes generating, in a reinforcement learning module, a counter-radar maneuver based on the operational model in response to classifying the radar network as non-friendly. Embodiments of the disclosure implement the counter-radar maneuver via the autonomous agent in communication with the reinforcement learning module.