Ego Motion Determination via Radar Map Accumulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining ego motion in vehicles lack accuracy and redundancy, particularly in autonomous systems, leading to potential safety and autonomy issues.
Innovation Solution
A computer-implemented method using radar data from multiple frames to derive ego motion by merging environment maps, computing motion vectors, and extracting a mask map with weights, which is processed by neural networks to enhance accuracy and filter relevant data, thereby improving vehicle safety and autonomy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current methods for determining ego motion are used, then the system is simple, but the accuracy and reliability are insufficient
Solution Approach 1:
The method segments the ego motion determination process into distinct computational stages: environment map generation from radar data, accumulation of multiple environment maps, motion vector field computation, and mask map extraction. This segmentation allows each stage to be optimized independently while maintaining overall system manageability despite increased complexity.
Solution Approach 2:
The patent transitions from traditional 2D environment maps to 3D accumulated maps with temporal dimension, enabling more accurate ego motion determination. The addition of the temporal dimension through map accumulation and the extraction of motion vectors across multiple frames provides redundant information that improves measurement precision.
2Measurement precision
If multiple frames of radar data are processed, then accuracy improves, but processing time increases
Solution Approach 1:
The method performs preliminary actions by pre-processing radar data into environment maps before accumulation. This preliminary transformation organizes the data in a format that facilitates efficient subsequent processing, reducing the computational burden when multiple frames are processed for improved accuracy.
Solution Approach 2:
The patent creates multiple copies of environment maps from different time frames and accumulates them. This copying approach allows parallel processing of multiple frames without sequentially analyzing raw radar data each time, thereby reducing overall processing time while maintaining the accuracy benefits of multi-frame processing.
3Reliability
If Doppler channels are reduced, then redundancy is incorporated, but speed information is lost
Solution Approach 1:
The method extracts speed information from the Doppler channels before reducing the number of channels. By separating and extracting the essential speed data, the system can then reduce Doppler channel count for redundancy purposes without permanently losing the extracted speed information, which is preserved for ego motion determination.
Solution Approach 2:
The patent uses motion vectors as an intermediary to transfer speed information from the reduced Doppler channels to the final ego motion calculation. This intermediary representation preserves the essential speed data while allowing the system to work with fewer Doppler channels, achieving both redundancy and information preservation.
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
The method provides high-accuracy ego motion determination, increasing vehicle safety and autonomy by filtering relevant motion vectors and incorporating redundancy through Doppler channel reduction and neural network processing.
Implementation Method 1
the vehicle having at least one radar emitter with a plurality of reception antennae
Implementation Method 2
The radar data includes a plurality of Doppler channels
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A computer implemented method to determine ego motion of a vehicle, the vehicle having at least one radar emitter with a plurality of reception antennae, the method including the steps of: • acquiring, from the reception antennae, different frames of radar data of the vehicle surrounding environment, each frame being acquired at a different time; • deriving from the radar data of each different frame, an environment map of the vehicle surrounding environment; and • deriving the ego motion of the vehicle by: - merging environment maps from at least two different frames into one accumulated map, - computing, from the accumulated map, a motion vector for each pixel of the accumulated map, - extracting, from the accumulated map, a mask map including a tensor mapping a weight for each pixel of the accumulated map.