20240010223. MULTIPLE SENSOR CALIBRATION IN AUTONOMOUS VEHICLES PERFORMED IN AN UNDEFINED ENVIRONMENT simplified abstract (GM Cruise Holdings LLC)
MULTIPLE SENSOR CALIBRATION IN AUTONOMOUS VEHICLES PERFORMED IN AN UNDEFINED ENVIRONMENT
Organization Name
Inventor(s)
Juan Fasola of San Francisco CA (US)
Ankit Rohatgi of Pacifica CA (US)
Zhonghua Ma of San Jose CA (US)
MULTIPLE SENSOR CALIBRATION IN AUTONOMOUS VEHICLES PERFORMED IN AN UNDEFINED ENVIRONMENT - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240010223 titled 'MULTIPLE SENSOR CALIBRATION IN AUTONOMOUS VEHICLES PERFORMED IN AN UNDEFINED ENVIRONMENT
Simplified Explanation
The abstract of the patent application describes a method for calibrating multiple sensors of an autonomous vehicle (AV) in an undefined training area. The AV is instructed to pilot itself along a path in the training area, and at a location where the AV has overlapped at least a portion of the path, the method determines if first returns from a previous lidar scan overlap with second returns from a subsequent lidar scan taken when the AV has overlapped the portion of the path. Initially, the AV does not have a location reference for objects in the training area, and its sensors are uncalibrated with respect to each other.
- The method involves instructing an AV to pilot itself in an undefined training area.
- The AV is instructed to follow a path until it overlaps at least a portion of the path.
- At the location of overlap, the method checks if lidar scans from before and after the overlap align.
- The AV initially lacks a location reference for objects in the training area.
- The sensors of the AV are uncalibrated with respect to each other.
Potential applications of this technology:
- Autonomous vehicles: This method can be used to calibrate sensors in autonomous vehicles, improving their accuracy and reliability.
- Training areas: The method can be employed in training areas to test and optimize the performance of autonomous vehicles.
Problems solved by this technology:
- Sensor calibration: The method addresses the challenge of calibrating multiple sensors in an autonomous vehicle, ensuring accurate and consistent data collection.
- Localization: By determining the overlap of lidar scans, the method helps establish a location reference for objects in the training area.
Benefits of this technology:
- Improved accuracy: Calibrating the sensors of an AV enhances the accuracy of data collected, leading to better decision-making and performance.
- Reliable localization: By establishing a location reference, the method enables precise localization of objects in the training area, contributing to safer navigation and operation of autonomous vehicles.
Original Abstract Submitted
the subject technology is related to autonomous vehicles (av) and, in particular, to calibrating multiple sensors of an av in an undefined training area. an example method includes instructing the av to pilot itself in an undefined training area subject to at least one constraint, wherein the av is instructed to pilot itself along a path until the av has overlapped at least a portion of the path, and at a location at which the av has overlapped at least the portion of the path, determining that first returns from a previous lidar scan overlaps with second returns from a subsequent lidar scan taken when the av has overlapped at least the portion of the path. initially, the av does not include a location reference to identify locations of objects in the undefined training area and a plurality of sensors of the av are uncalibrated with respect to each other.
