Map creation and localization for autonomous driving applications

US11698272B2 · kind B2 · utility

Assignee

• NVIDIA Corporation · US

Inventors

• Michael Kroepfl · Bellevue, US
• Amir Akbarzadeh · San Francisco, US
• Ruchi Bhargava · Redmond, US
• Vaibhav Thukral · Kirkland, US
• Neda Cvijetic · East Palo Alto, US
• Vadim Cugunovs · Seattle, US
• David Nister · Bellevue, US
• Birgit Henke · Seattle, US
• Ibrahim Eden · Redmond, US
• Youding Zhu · Columbus, US
• Michael Grabner · Redmond, US
• Ivana Stojanovic · Palo Alto, US
• Yu Sheng · San Diego, US
• Jeffrey Liu · Bellevue, US
• Enliang Zheng · Redmond, US
• Jordan Marr · Santa Clara, US
• Andrew Carley · Kenmore, US

Key dates

• Filing date: Aug 31, 2020
• Grant date: Jul 11, 2023
• Expiry date: Apr 2, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N3/063
• WIPO field: Measurement
• WIPO sector: Instruments

Abstract

An end-to-end system for data generation, map creation using the generated data, and localization to the created map is disclosed. Mapstreams—or streams of sensor data, perception outputs from deep neural networks (DNNs), and/or relative trajectory data—corresponding to any number of drives by any number of vehicles may be generated and uploaded to the cloud. The mapstreams may be used to generate map data—and ultimately a fused high definition (HD) map—that represents data generated over a plurality of drives. When localizing to the fused HD map, individual localization results may be generated based on comparisons of real-time data from a sensor modality to map data corresponding to the same sensor modality. This process may be repeated for any number of sensor modalities and the results may be fused together to determine a final fused localization result.