Pony.ai has been granted a patent for a computer-implemented method and system for training an autonomous driving model. The method involves creating time-dependent 3D traffic environment data using real or simulated traffic element data, and then using a generative adversarial network (GAN) model to create simulated time-dependent 3D traffic environmental data. This data is used to train a computer-based autonomous driving model, which is then used for virtual driving operations with dynamically changing routes. GlobalData’s report on Pony.ai gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on Pony.ai, Autonomous freight management was a key innovation area identified from patents. Pony.ai's grant share as of September 2023 was 50%. Grant share is based on the ratio of number of grants to total number of patents.
Training autonomous driving model using simulated 3d traffic data
A recently granted patent (Publication Number: US11774978B2) describes a computer-implemented method and system for creating and using simulated time-dependent 3D traffic environmental data to train a computer-based autonomous driving model. The method involves creating time-dependent 3D traffic environment data using real or simulated traffic element data. This data is then used to generate simulated time-dependent 3D traffic environmental data by applying a time-dependent 3D generative adversarial network (GAN) model. The simulated data includes various elements such as vehicular sounds, noises from pedestrians, density of pedestrians, movements or noises of plants, level of brightness, walking patterns of pedestrians, vibration of vehicles attributed to road conditions, and sounds from humans.
The simulated time-dependent 3D traffic environmental data is used to train a computer-based autonomous driving model. The method also involves executing a virtual driving operation based on the simulated data. This virtual driving operation includes traversing one or more intermediate check points and selecting a dynamically changing route based on driving mode, cost, and the difference between planned and actual operations. The dynamically changing route must pass through the intermediate check points.
The system described in the patent includes one or more processors and memory storing instructions for executing the method. The system creates time-dependent 3D traffic environment data and generates simulated time-dependent 3D traffic environmental data using a time-dependent 3D GAN model. The simulated data includes elements such as vehicular sounds, noises from pedestrians, density of pedestrians, movements or noises of plants, level of brightness, walking patterns of pedestrians, vibration of vehicles attributed to road conditions, and sounds from humans. The simulated data is used to train a computer-based autonomous driving model, and the system executes a virtual driving operation based on the simulated data, including traversing intermediate check points and selecting dynamically changing routes.
This patent presents a method and system for generating realistic and time-dependent 3D traffic environmental data to train autonomous driving models. By incorporating various elements such as sounds, pedestrian behavior, plant movements, and road conditions, the simulated data aims to provide a comprehensive and accurate representation of real-world traffic environments. The virtual driving operation allows for testing and evaluating the performance of autonomous driving models in different scenarios. This technology has the potential to enhance the development and testing of autonomous vehicles, contributing to the advancement of self-driving technology.
To know more about GlobalData’s detailed insights on Pony.ai, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
