Vehicle Radar Mapping and Localization Without GPS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radar systems in autonomous vehicles face challenges in accurately determining location and navigating without relying on GPS or other location services, particularly when operating in environments with varying terrain or obstructions, as they lack the resolution and data processing capabilities to effectively correlate radar data with map information in real-time.
Innovation Solution
The implementation of a radar system that transmits and receives multiple pulses while in motion, utilizing Synthetic Aperture Radar (SAR) mode to enhance resolution and process signals from different angles, allowing for the correlation of radar data with map information to determine vehicle location and control autonomous navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radar systems transmit and receive multiple signal pulses while in motion to enhance location determination, then measurement precision of vehicle location is improved, but device complexity increases due to need for SAR processing and signal correlation
Solution Approach 1:
The system pre-processes radar reflection signals by correlating them with predetermined target information during data collection phases, preparing processed data structures that can be quickly queried during navigation. This preliminary correlation of radar data with map information reduces the computational burden during real-time location determination.
Solution Approach 2:
The system creates synthetic aperture representations of the physical radar aperture by processing multiple pulses received at different vehicle locations. These synthetic aperture images are stored as processed data that can be referenced without re-processing original raw signals, enabling efficient location determination through data correlation rather than repeated complex signal processing.
2Measurement precision
If radar systems process signals from multiple angles using SAR mode to improve resolution, then measurement precision of environmental features is improved, but loss of time increases due to extended signal processing requirements
Solution Approach 1:
The system performs correlation processing of radar signals with predetermined target information during data collection and storage phases, pre-computing the relationship between radar reflections and map features. This allows real-time location determination to simply query pre-processed correlations rather than perform full SAR processing, dramatically reducing computation time while maintaining high resolution.
Solution Approach 2:
The system dynamically adjusts the level of processing applied to radar signals based on operational context. During mapping phases, full SAR processing with multiple pulses is applied to build high-resolution databases. During navigation phases, the system uses pre-processed correlations and fewer pulses, dynamically optimizing the balance between resolution and processing time based on current needs.
3Reliability
If radar systems correlate target information with predetermined map data to determine location, then reliability of autonomous navigation is improved, but device complexity increases due to integration of mapping and localization functions
Solution Approach 1:
The radar system performs multiple functions using the same hardware and processing pipeline: it can operate in mapping mode to build environmental databases, in localization mode to determine vehicle position, and in obstacle detection mode for safety. This multi-functionality reduces overall system complexity compared to having separate dedicated systems for each function, while maintaining high reliability through consistent data processing.
Solution Approach 2:
The system merges mapping and localization functions into a unified processing framework where radar reflection data is correlated with predetermined map information to simultaneously achieve environmental understanding and self-localization. This integration allows the system to cross-validate data from multiple sources, improving reliability while reducing the complexity of managing separate independent systems.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables improved location determination and navigation in autonomous vehicles by enhancing radar resolution and data processing, allowing for accurate correlation of radar data with map information, even in complex environments, thereby enabling reliable autonomous operation without reliance on GPS.
Implementation Method 1
Radio detection and ranging (RADAR) systems can be used to actively estimate distances to environmental features by emitting radio signals and detecting returning reflected signals
Implementation Method 2
Some systems may also estimate relative motion of reflective objects based on Doppler frequency shifts in the received reflected signals
Implementation Method 3
The radar system may be configured to transmit a radio waveform that that include a linear frequency modulation (LFM)
Data Source
AI summary
In an example method, a vehicle configured to operate in an autonomous mode could have a radar system used to aid in vehicle guidance. The method could include transmitting at least two signal pulses. The method further includes, for each transmitted signal pulse, receiving a reflection signal associated with reflection of the respective transmitted signal pulse. Each reflection signal may be received when the apparatus is in a different respective location. Additionally, the method includes processing the received reflection signals to determine target information relating to one or more targets in an environment of the vehicle. Also, the method includes correlating the target information with at least one object of a predetermined map of the environment of the vehicle to provide correlated target information. Yet further, the method includes storing the correlated target information for the at least one object in an electronic database.