Autonomous Vehicle Sensor Verification Against Spoofed Map Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face safety risks due to inaccurate sensor information caused by malicious sensor feeds, which can lead to unsafe maneuvers such as sudden stopping or incorrect navigation, as attackers can broadcast fake signals that confuse radar or lidar sensors.
Innovation Solution
An autonomous vehicle sensor security system that generates and shares 'map instances' of environmental information securely between vehicles and roadside equipment, using Transport Layer Security (TLS) for authentication, and activates a secure sensor only when discrepancies are detected to verify the accuracy of the environment, thereby minimizing the impact of malicious signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a secure sensor is activated continuously to verify environmental data, then the reliability of sensor information is improved, but the energy consumption increases
Solution Approach 1:
The secure sensor is activated periodically or on-demand rather than continuously. Specifically, the secure sensor is activated when discrepancies are detected between map instances from different autonomous vehicles, allowing verification of environmental data only when necessary. This periodic activation maintains reliability improvement while significantly reducing energy consumption compared to continuous operation.
2Reliability
If multiple map instances are correlated and verified between vehicles, then the security against malicious sensor feeds is improved, but the device complexity increases
Solution Approach 1:
The verification system is segmented into distributed components where each autonomous vehicle independently generates and shares its own map instances. The correlation and verification process is distributed across multiple vehicles rather than centralized in one complex system. Each vehicle performs local comparison of received map instances against its own map instance, reducing individual device complexity while achieving collective security improvement.
3Measurement precision
If a secure sensor is deployed to verify discrepancies, then the measurement precision of environmental data is improved, but the loss of time for verification increases
Solution Approach 1:
Map instances are generated and shared in advance before discrepancies are detected. Autonomous vehicles continuously generate and exchange map instances representing their environmental perceptions. When a discrepancy is detected, the secure sensor verifies the issue using pre-shared map instances from other vehicles, eliminating the need for real-time verification and reducing the time penalty for accuracy improvement.
Data Source
AI summary
Example methods and systems are disclosed to provide autonomous vehicle sensor security. An example method may include generating, by a first autonomous vehicle, a first map instance of a physical environment using first environmental information generated by a first sensor of a first autonomous vehicle. A second map instance from at least one of a second autonomous vehicle located in the physical environment is received. The first map instance may be correlated with the second map instance. In response to a discrepancy between the first map instance and the second map instance, a secure sensor may be activated to generate a third map instance. In response to the third map instance verifying that the discrepancy accurately describes the physical environment, the first environmental information including the discrepancy is used to navigate the first autonomous vehicle.