20230113286. CREATING MULTI-RETURN MAP DATA FROM SINGLE RETURN LIDAR DATA simplified abstract (HERE GLOBAL B.V.)
CREATING MULTI-RETURN MAP DATA FROM SINGLE RETURN LIDAR DATA
Organization Name
Inventor(s)
Deekshant Saxena of Mumbai (IN)
CREATING MULTI-RETURN MAP DATA FROM SINGLE RETURN LIDAR DATA - A simplified explanation of the abstract
This abstract first appeared for US patent application 20230113286 titled 'CREATING MULTI-RETURN MAP DATA FROM SINGLE RETURN LIDAR DATA
Simplified Explanation
The patent application describes a system and methods for creating multi-return map data using single return lidar data. The system utilizes a combination of a long short-term memory (LSTM) model and a generative adversarial network (GAN) model.
- The system uses a single return of lidar data at a specific time stamp and generates multiple unseen samples of second and third returns.
- An LSTM model is employed to create a sequential calibration based on the incidence angle, allowing the system to choose the optimized second and third returns at the same time stamp.
- This approach effectively creates a localized model of three returns from a single return of lidar, providing additional data to generate a high-definition (HD) map.
Potential Applications
- Autonomous vehicles: The technology can be used to enhance the perception capabilities of autonomous vehicles by providing more accurate and detailed map data.
- Urban planning: The system can assist in creating detailed maps for urban planning purposes, helping to optimize infrastructure development and city design.
- Environmental monitoring: By generating multi-return map data, the technology can aid in monitoring and analyzing environmental changes, such as vegetation growth or land erosion.
Problems Solved
- Limited lidar data: Traditional lidar systems often provide only a single return, limiting the amount of information that can be extracted from the environment.
- Incomplete mapping: Without multiple returns, the accuracy and completeness of generated maps may be compromised, leading to potential errors in navigation and perception tasks.
Benefits
- Enhanced map data: By generating multiple unseen samples of second and third returns, the system improves the quality and richness of map data, enabling more precise localization and object detection.
- Cost-effective solution: The technology leverages existing single return lidar data, eliminating the need for additional sensors or hardware upgrades.
- Improved perception capabilities: The localized model of three returns allows for better understanding of the environment, enhancing the perception capabilities of autonomous systems and aiding in decision-making processes.
Original Abstract Submitted
system and methods for creating multi-return map data using single return lidar data. the systems and methods use a long short-term memory (lstm) model in combination with a generative adversarial network (gan) model. the systems and method use a single (1st) return of lidar at a time stamp and create multiple unseen samples of 2nd and 3rd returns. the lstm model is used to create a sequential calibration based on incidence angle to choose the optimized 2nd and 3rd return at the same instance of the time stamp. this creates a localized model of three returns from a single return of lidar and thus provides additional data to generate an hd map.
