Earth-Frame Vehicle Mapping for Absolute Object Velocity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing algorithms for determining absolute object velocities from a vehicle's sensor data are computationally complex and struggle to differentiate between static and dynamic objects, as they are based on the vehicle's frame of reference.
Innovation Solution
A method that calculates absolute object velocities using a grid defined in the earth frame of reference, updating occupancy values from sensor data, and applying motion detection algorithms to distinguish between static and dynamic objects, simplifying the computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing algorithms are used to determine absolute object velocities from vehicle sensor data, then object velocity measurement is achieved, but computational complexity increases significantly
Solution Approach 1:
The patent inverts the conventional approach by transforming sensor data from the vehicle's frame of reference to the earth's frame of reference. Instead of calculating absolute velocities directly from relative vehicle-centered data using complex algorithms, the system first transforms occupancy grids into an earth-fixed coordinate system where static objects have zero velocity, thereby simplifying the velocity calculation process while maintaining measurement precision
2Device complexity
If simple motion detection algorithms are used to measure occupancy change between two time-instances, then computational complexity is reduced, but the ability to differentiate between static and dynamic objects is lost
Solution Approach 1:
The patent applies coordinate frame inversion to enable simple motion detection algorithms to work effectively. By transforming the occupancy grid from the vehicle's moving frame of reference to the earth's stationary frame of reference, static objects naturally appear as stationary (zero velocity) in the transformed data, allowing simple algorithms to correctly differentiate between static and dynamic objects without requiring complex computational logic
3Ease of operation
If sensor data is processed in the vehicle's frame of reference, then data processing is simplified, but differentiation between static and dynamic objects becomes difficult
Solution Approach 1:
The patent resolves this contradiction by inverting the reference frame from vehicle-centered to earth-centered. In the earth's frame of reference, the vehicle becomes the moving element while the environment becomes stationary, causing static objects to have zero velocity in the transformed occupancy grid. This inversion maintains processing simplicity while enabling accurate static-dynamic object differentiation through straightforward velocity thresholding
Data Source
AI summary
Provided is a computer-implemented method for mapping a vehicle environment, the method comprising: a) defining a first grid in the earth frame of reference having a first coordinate system; b) initializing a position of the vehicle in the first grid and a first set of cell occupancy values of the first grid; c) receiving sensor data of the surroundings of the vehicle from one or more sensors on the vehicle; d) updating the first grid with a second set of occupancy values calculated at least in part from the sensor data; and e) calculating one or more absolute velocities of one or more objects in the earth frame of reference from the change in cell occupancy values of the first grid.