Auto-Panning Camera Mirror Using Weighted Trailer Angle Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing camera mirror systems for commercial vehicles struggle with inaccurate trailer angle estimates during reversing maneuvers, leading to insufficient view adjustments and frequent manual intervention, which can disrupt operations.
Innovation Solution
A method and system that utilize multiple trailer angle estimation methods, assign confidence values, calculate a weighted sum, and apply low-pass filtering to provide a reliable and smooth automatic panning of the camera view, ensuring the trailer's rear remains visible by fusing various detection systems' outputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple trailer angle estimation methods are used with confidence values and weighted sums, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent combines multiple trailer angle estimation methods (image-based detection, sensor-based detection, and kinematic model-based estimation) into a unified system that fuses their outputs using confidence values and weighted sums. This merging approach leverages the strengths of each method while compensating for their individual weaknesses, thereby improving overall measurement precision despite the increased system complexity.
Solution Approach 2:
The patent introduces confidence values as an intermediary mechanism to mediate between multiple estimation methods. Each estimation method's output is assigned a confidence value that reflects its reliability under current operating conditions. These confidence values serve as weights in the fusion process, allowing the system to dynamically adjust the contribution of each estimation method based on its current accuracy, thus improving overall precision while managing complexity through a structured fusion approach.
2Ease of operation
If automatic panning is implemented based on trailer angle estimates, then ease of operation is improved, but reliability deteriorates due to inaccurate angle estimates
Solution Approach 1:
The patent implements a feedback mechanism where the automatic panning system continuously receives updated trailer angle estimates from multiple sources, fuses them with confidence values, and adjusts the camera view accordingly. The system monitors the accuracy and consistency of angle estimates in real-time, and when confidence values indicate unreliable estimates, it can reduce the aggressiveness of panning adjustments or switch to manual control mode, thereby maintaining reliability while preserving ease of operation.
Solution Approach 2:
The patent makes the automatic panning system dynamic by allowing it to adapt its behavior based on the confidence values of trailer angle estimates. When confidence is high, the system performs aggressive automatic adjustments for ease of operation. When confidence is low, the system reduces automatic intervention and allows more manual control, thus dynamically balancing ease of operation with reliability based on current estimation quality.
3Adaptability or versatility
If manual panning adjustments are allowed, then adaptability is improved, but productivity decreases due to frequent stopping
Solution Approach 1:
The patent implements a self-service automatic panning system that autonomously adjusts the camera view based on fused trailer angle estimates without requiring operator intervention. The system automatically detects trailer angle changes, calculates appropriate pan adjustments, and executes view changes continuously during maneuvers. This eliminates the need for operators to stop and manually adjust views, thereby maintaining adaptability through automated decision-making while significantly improving productivity by ensuring continuous maneuver execution.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A method for automatically panning a view for a commercial vehicle includes determining a plurality of estimated trailer angles. Each estimated trailer angle is determined using a distinct estimation method, and method assigns a confidence value to each estimated trailer angle in the plurality of estimated trailer angles. The method determines a weighted sum of the plurality of estimate trailer angles, and automatically pans the view based at least in part on the weighted sum and a current vehicle operation.