Apple adds Kalman filter to its mapping system to provide vehicles with more accurate map information

According to foreign media reports, on November 9th local time, the US Patent Office (US Patent Office) announced a patent application from Apple. The patent is related to estimating the location and/or direction of the device, which can provide users with accurate map information on devices such as iPhones and on vehicles equipped with Apple's in-vehicle system CarPlay (cars, trucks, motorcycles, watercraft, airplanes, etc.), thereby more accurately tracking the movement of people and vehicles.

(Image source: patentlyapple.com)

Apple's system will use traditional technologies such as GPS/GNSS, visual inertial odometry modules and inertial measurement units (IMUs) for its mapping system, but will also adopt a new technology - the Kalman Filter.

"In one or more instances, the architecture may improve estimates of a device's location, such as for use in augmented reality applications such as Apple Maps and Maps on CarPlay," Apple's patent application reads.

The improved estimated information may be used to present digital content (such as visual, audio, and/or tactile feedback) in an image of a real environment (such as an image captured by a sensor). For example, notification information (such as prompt information, overlay information, audio prompts, tactile feedback, etc.) about time and/or location may be provided based on the estimated information of the architecture.

(Image source: patentlyapple.com)

In an example architecture, a GNSS receiver and a visual inertial odometry module may be included as shown in patent FIG. 3 to estimate the position and/or orientation of the device.

(Image source: patentlyapple.com)

In addition, the architecture may include an extended Kalman filter. As shown in patent FIG. 5, the Kalman filter may receive signals from the GNSS receiver, and the visual inertial odometry module may be an input device and may also output estimated position information of the device.