Tesla Patents

Autonomous Driving

Tesla filed these patents around Autonomous Driving in the last 5 years

# Patent No. Short Description
1. US20210271259A1 A technique for generating training data for deep learning models using existing models and sensors on vehicles
2. US10406972B2 Automatically activating a turn signal source in a vehicle when it will imminently leave its current lane by either crossing over a lane line or effecting a turn
3. US11748620B2 A technique to train machine learning models for autonomous driving using time series data from vehicle sensors like cameras
4. US20180188733A1 Using machine learning to simulate high-capacity sensor data from lower capacity sensors for autonomous vehicle control
5. US20230274591A1 A vehicular autonomous driving system that predicts potential future component failures in the autonomous driving system
6. US10534092B2 Using vehicles as reference stations to improve positioning accuracy through sharing offsets and raw data
7. US11816585B2 Improving object detection in autonomous vehicles by leveraging multiple image processing engines with different speeds and accuracies
8. US20200348909A1 Training machine learning models for autonomous and driver assisted driving using a scalable training platform with multiple node engines that can be arranged in a network
9. US11537811B2 Object detection method using a cropped high resolution field of view (FOV) along with a downsampled low resolution image to improve accuracy and range of object detection
10. US11215999B2 Pre-processing sensor data from vehicles to improve the accuracy of deep learning models used for autonomous driving
11. US11150664B2 A technique for generating highly accurate machine learning results for autonomous driving using training data created from time series of captured vehicle sensor data
12. US20230176593A1 Automatically summoning a vehicle to a user's location using sensor data and machine learning to generate a representation of the environment around the vehicle
13. US11636333B2 Model training and implementation pipeline for autonomous control systems that leverages model performance estimation and selection to generate and deploy machine-learned models efficiently
14. US20220284712A1 Training a machine learning model to accurately predict object properties like distance and direction using only vision data from a vehicle's cameras
15. US10678244B2 Synthesizing sensor data for autonomous vehicle control systems to improve training of computer models

Download the full patent report containing all of the 24 patents