Radar Velocity Profiling for Faster, More Accurate Object Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately forecasting the movement of objects due to the limitations of existing sensor data processing methods, which either provide rapid but inaccurate short-term analysis or long-term tracking without rapid change detection.
Innovation Solution
A method that combines top-down and bottom-up sensor data processing by integrating radar and LiDAR data to generate enhanced velocity profiles using Kalman filtering, allowing for accurate forecasting of object movement by correlating historical and instantaneous data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If top-down sensor data processing is used, then rapid change detection is improved, but measurement precision deteriorates
Solution Approach 1:
The patent combines top-down and bottom-up sensor data processing approaches into a unified system. The top-down approach processes radar data for rapid change detection while the bottom-up approach processes LiDAR data for precise trajectory forecasting. By merging these complementary approaches, the system achieves both rapid detection speed and high measurement precision that neither approach could achieve alone.
2Measurement precision
If bottom-up sensor data processing is used, then measurement precision is improved, but speed deteriorates
Solution Approach 1:
The system integrates bottom-up LiDAR processing for precise trajectory prediction with top-down radar processing for rapid change detection. The LiDAR data provides high-resolution spatial information for accurate forecasting, while radar data quickly identifies changes. This merging allows the system to maintain high measurement precision while achieving rapid response to dynamic changes.
Solution Approach 2:
The patent uses an intermediary data fusion mechanism that combines processed results from both top-down and bottom-up approaches. This intermediary layer integrates the rapid detections from radar with the precise trajectories from LiDAR, mediating between the two processing streams to produce unified, accurate, and timely object trajectory predictions.
3Device complexity
If single sensor type processing is used, then device complexity is reduced, but reliability deteriorates
Solution Approach 1:
The patent merges data from multiple sensor types (radar and LiDAR) with different processing approaches (top-down and bottom-up) into a unified forecasting system. This combination of diverse sensor data sources and processing methods enhances the reliability and robustness of object trajectory predictions, compensating for the increased device complexity through systematic integration.
Data Source
AI summary
The present disclosure is directed to combining the strengths of different methods of analyzing collected sensor data to reduce the risk of an autonomous vehicle (AV) impacting an object. This may include combining data from sets of data that track movement of objects over time with instantaneously received sensor data based on a series of steps that include accessing data that tracks the motion of objects in the field of view of a sensing apparatus, receiving current sensor data that includes a component of current or instantaneous object motion, and generating a forecast of future motion of that object. This forecast may be based on an analysis that compares the data that tracks the motion of objects with the current sensor data as part of a process that generates a risk probability. When the risk probability meets or exceeds a threshold level, a corrective action may be initiated.