Mobile Object Motion Surface Estimation via 3D Point Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately estimating the surface for motion of a mobile object, such as a vehicle, which affects the precision of obstacle detection and movement path calculation, especially when the object's orientation and speed vary significantly.
Innovation Solution
An estimation device that performs three-dimensional measurement of the surroundings, calculates movement information, sets a threshold value based on time series differences, extracts relevant three-dimensional points, divides space, selects representative points, and estimates the surface for motion, enhancing accuracy by approximating these points to determine the surface and detect obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If three-dimensional measurement is performed using a camera to estimate the surface for motion, then the measurement coverage is improved, but the estimation accuracy deteriorates when the mobile object's orientation and speed vary significantly
Solution Approach 1:
The patent divides the three-dimensional measurement data into multiple time-series groups corresponding to different movement states (different orientations and speeds). By segmenting the data based on movement conditions, the system can selectively process measurements appropriate for each state, thereby maintaining accuracy despite variations in motion parameters.
Solution Approach 2:
The patent pre-establishes multiple surface estimation models corresponding to different movement states (various orientations and speeds). Before actual surface estimation, the system determines the current movement state and selects the appropriate pre-prepared model, avoiding the need to recalculate from scratch and maintaining consistency and accuracy across different motion conditions.
2Device complexity
If a single surface estimation model is used for all movement conditions, then the device complexity is reduced, but the estimation accuracy deteriorates under varying orientation and speed
Solution Approach 1:
The patent implements a dynamic model selection mechanism that automatically switches between different surface estimation models based on real-time movement state detection. The system monitors orientation and speed parameters, and dynamically selects the most appropriate pre-prepared model for current conditions, achieving high accuracy without requiring a single complex universal model.
Solution Approach 2:
The patent changes the selection criterion from a fixed single-model approach to a parameter-based multi-model approach. By using movement state parameters (orientation angle, speed) as selection criteria, the system chooses the optimal model for current conditions, thereby maintaining accuracy while managing complexity through parameter-driven model selection.
Data Source
AI summary
According to an embodiment, a device includes processing circuitry. The processing circuitry performs three-dimensional measurement of surroundings of a mobile object to obtain a three-dimensional point group. The processing circuitry calculates movement information. The processing circuitry makes a threshold value based on time series differences in the movement information. The processing circuitry extracts, from the three-dimensional group, three-dimensional points having distance to the mobile object in a moving direction of the mobile object to be equal to or smaller than the threshold value. The processing circuitry divides space, in which the three-dimensional points are present, into divided spaces in the moving direction. The processing circuitry selects a representative point for the divided space from among three-dimensional points included in the divided space. The processing circuitry estimates, as a surface for motion on which the mobile object moves, a surface which approximates the representative points.