Autonomous Vehicle Object Representation Smoothing Using Map Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in generating realistic representations of detected objects using laser sensor data, leading to unrealistic shifting and rotating of object representations, which can be disconcerting for passengers and potentially dangerous if smoothing techniques compromise accuracy for vehicle control.
Innovation Solution
The method involves using pre-stored map information to smooth laser data points before generating object representations, adjusting characteristics such as position, orientation, and size based on proximity to lane or parking spot center points, to create a smooth and accurate visual representation without affecting vehicle control data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is used directly to generate object representations, then the representations are accurate for vehicle control, but the representations exhibit unrealistic shifting and rotating that disconcerts passengers
Solution Approach 1:
The system segments the object representation into two distinct components: a first set of characteristics (position, orientation, size) derived from sensor data for control accuracy, and a second set of characteristics derived from map data for visual stability. This segmentation allows each component to serve its specific function without compromising the other.
Solution Approach 2:
Map data serves as an intermediary between the sensor data and the final object representation displayed to passengers. The map data provides a stable reference framework that mediates the inherently noisy sensor measurements, producing visually stable representations while preserving the accuracy needed for control.
2Ease of operation
If smoothing techniques are applied to sensor data, then object representations become more stable visually, but the accuracy for vehicle control may be compromised
Solution Approach 1:
The system segments the characteristics used for control from those used for display. Control-critical characteristics are taken directly from sensor data without smoothing, while display characteristics are derived by combining sensor data with stable map data, achieving visual stability without sacrificing control accuracy.
Solution Approach 2:
Instead of smoothing the original sensor data, the system creates a copied representation for display purposes by integrating sensor measurements with map data. This copying approach provides visual stability while the original unsmoothed sensor data remains available for accurate vehicle control.
3Ease of operation
If map information is used to smooth sensor data, then realistic object representations are generated, but the system complexity increases
Solution Approach 1:
The system uses map data to serve multiple functions simultaneously: providing a stable reference frame for visual representation, defining geographic boundaries for context-aware display, and enabling consistent object tracking across sensor sweeps. This multi-functionality reduces the need for separate smoothing mechanisms.
Solution Approach 2:
The system changes the parameters used for different purposes: sensor data parameters (high precision, noisy) are used for control, while combined sensor-map parameters (smoothed, stable) are used for display. This parameter differentiation achieves realism without requiring complex smoothing algorithms.
Data Source
AI summary
The technology relates to generating a representation of an object detected by a perception system of a vehicle. For instance, a set of data points corresponding to an object is received from the perception system. Map information identifying shapes and locations of lanes as well as locations of center points for the lanes is retrieved. A representative point of the center points is selected based on the set of data points. A first position is determined based on the location of the representative point. A second position is determined based on the set of data points. The first and second positions are used to determine a third position for a representation of the object based on a distance the first position and the second position. The he representation of the object is generated using the third position. The representation of the object is displayed on a display of the vehicle