AGV Orientation Sensor Validation Using Fixed-Object Vision Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous ground vehicles (AGVs) face challenges in accurately determining their orientation due to malfunctioning or worn-out orientation sensors, which can lead to unintended behavior and potential damage, as existing validation methods are inadequate for timely diagnosis.
Innovation Solution
A method and system that validate orientation sensor readings by receiving distance and angle data of fixed objects from a vision sensor, calculating orientations at different positions, determining the actual change in orientation, and comparing it with sensor readings to validate the sensor's accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If orientation sensor readings are used for navigation, then the AGV can determine its orientation, but the sensor may malfunction or provide inappropriate results due to wear and tear
Solution Approach 1:
The system continuously monitors orientation sensor readings by comparing them against orientation values calculated from vision sensor data. This feedback mechanism allows the system to detect when the orientation sensor provides inappropriate results due to malfunction or wear, enabling timely diagnosis and preventing unintended AGV behavior.
Solution Approach 2:
The vision sensor acts as an intermediary to validate orientation sensor readings. By calculating orientation independently using vision sensor data and comparing it with orientation sensor readings, the system can identify sensor failures without relying solely on the potentially faulty orientation sensor.
2Measurement precision
If validation methods are implemented to diagnose sensor failures, then sensor accuracy can be verified, but the system complexity increases
Solution Approach 1:
The system uses its existing vision sensor, which is already mounted for other navigation purposes, to validate the orientation sensor. This self-service approach allows the system to perform validation without adding dedicated validation hardware, thereby limiting the increase in system complexity while still achieving accurate orientation verification.
3Device complexity
If the AGV relies on a single orientation sensor, then the system remains simple, but the ability to quickly and accurately determine orientation decreases when the sensor fails
Solution Approach 1:
The system continuously calculates orientation from vision sensor data in advance and compares it with orientation sensor readings in real-time. This preliminary calculation ensures that when the orientation sensor fails, the system can immediately switch to using the pre-calculated orientation from vision data, maintaining quick and accurate orientation determination without requiring complex redundant sensor systems.
Data Source
AI summary
This disclosure relates to method and system for validating readings of orientation sensor mounted on autonomous ground vehicle (AGV). The method may include receiving distances and angles of observation of at least two fixed objects with respect to AGV at a first position and then at a second position, calculating a first orientation and a second orientation of AGV at the first position and at the second position respectively based on the distances, the angles of observation, and coordinate positions of each of the at least two fixed objects, determining an actual change in orientation of AGV based on the first orientation and the second orientation, and validating the readings of the orientation sensor based on the actual change in orientation. The at least two fixed objects are objects with pre-identified properties in a field of view of a vision sensor mounted on AGV and on both sides of AGV.