The present disclosure provides systems and methods that combine physics-based systems with machine learning to generate synthetic LiDAR data that accurately mimics a real-world LiDAR sensor system. In particular, aspects of the present disclosure combine physics-based rendering with machine-learned models such as deep neural networks to simulate both the geometry and intensity of the LiDAR sensor. As one example, a physics-based ray casting approach can be used on a three-dimensional map of an environment to generate an initial three-dimensional point cloud that mimics LiDAR data. According to an aspect of the present disclosure, a machine-learned model can predict one or more dropout probabilities for one or more of the points in the initial three-dimensional point cloud, thereby generating an adjusted three-dimensional point cloud which more realistically simulates real-world LiDAR data.

A LIDAR system includes one or more optical components that output multiple system output signals. The system also includes electronics that use light from the system output signals to generate LIDAR data. The LIDAR data indicates a distance and/or radial velocity between the LIDAR system and one or more object located outside of the LIDAR system. The electronics including a series processing component that processes electrical signals that are each generated from one of the system output signals. The series processing component processes the electrical signals generated from different system output signals in series.

A position management device includes: a sensor which acquires point cloud data in each unit time; a pair selection unit which selects a pair of the coordinates between first point cloud data and second point cloud data; and a position estimation unit which estimates a position of the moving object by making both coordinates of the pair selected by the pair selection unit correspond to each other and performing positioning processing of the first point cloud data and the second point cloud data. The pair selection unit selects the pair of the coordinates by pairing a first point belonging to the first point cloud data and a second point belonging to the second point cloud data in a case where a difference between a direction of the first point and a direction of the second point is equal to or smaller than a predetermined first threshold.

Disclosed herein are examples of ladar systems and methods where data about a plurality of ladar returns from prior ladar pulse shots gets stored in a spatial index that associates ladar return data with corresponding locations in a coordinate space to which the ladar return data pertain. This spatial index can then be accessed by a processor to retrieve ladar return data for locations in the coordinate space that are near a range point to be targeted by the ladar system with a new ladar pulse shot. This nearby prior ladar return data can then be analyzed by the ladar system to help define a parameter value for use by the ladar system with respect to the new ladar pulse shot. Examples of such adaptively controlled parameter values can include shot energy, receiver parameters, shot selection, camera settings, and others.

A method of automatically determining sensor placement in a target environment. The method comprises receiving as inputs: a three-dimensional (3D) map of the target environment; a number of a plurality of sensors to be placed in the target environment; a set of placement configuration parameters for the plurality of sensors; and a set of constraints for the plurality of sensors in the target environment. The method includes, based on the received inputs, using a simulation platform to simulate, based on one or more defined conditions in the target environment, operation of the plurality of sensors to generate a dataset comprising simulated output data for the plurality of sensors; and using a Reinforcement Learning (RL) algorithm to determine from the dataset comprising simulated output data for the plurality of sensors an optimized or at least an improved set of placement configuration parameters for the plurality of sensors.

Described herein is a method for beam profile analysis using at least one camera. The method includes: a) at least one data acquisition step; b) at least one image compression step; and c) at least one evaluation step.

A photomultiplier tube protection system with dual optical receiving channels for bathymetry LiDAR is designed, through a photomultiplier tube gating technology, based on dual optical receiving channels, main control module with STM32 single chip microcomputer and, high-speed AD sampling module. The system includes: calculating laser echo receiving power ratios of different optical receiving channels, respectively; acquiring, by AD sampling module, laser echo signal, and performing peak determination on acquired data, and transmitting peak information to the main control module; and collecting, by the main control module, echo signal intensity information, performing photomultiplier tube gating control according to the received echo signal intensity and the calculated echo receiving efficiency ratios of different optical receiving channels, and stopping the photomultiplier tube through photomultiplier tube gating control if saturated echo signal occurs, and adjusting external laser device power, thus achieving multiple protection of the photomultiplier tube.

A simulation system may generate radar data for synthetic simulations of autonomous vehicles, by using a data store of object radar data and attributes determined from sensor data captured in real-world physical environments. The radar data store may include radar point clouds representing real-world objects and associated object attributes, as well as radar background data captured for a number of physical environments. The simulation system may construct radar data for use in a simulation based on radar object data and/or radar background data, including using various probabilities within various overlay regions to determine subsets of object and background radar points to be rendered. During a simulation, the generated radar data may be provided to a simulated radar sensor of a simulated vehicle configured to execute trained perception models based on radar data input.

Simulating data received by a detection and ranging sensor including determining a first set of sample points on a surface of a simulated detector, the first set of sample points including a first sample point at a first location on the detector and a second sample point at a second location on the detector, generating a first ray associated with the first sample point, the first ray representing electromagnetic radiation reflected at a first point of intersection within a scene and incident on the first sample point on the surface of the detector; generating a second ray associated with the second sample point; the second ray representing electromagnetic radiation reflected at a second point of intersection within the scene and incident on the second sample point on the surface of the detector; and generating a ray-based data representation based on the first ray and the second ray.

A distance detection apparatus includes: an emitting light source, a receiving lens, a detection assembly, and a controller. The receiving lens and the emitting light source are spaced apart from each other, and the detection assembly is located near a focal plane of the receiving lens. The emitting light source is configured to emit detection light with a divergence angle to a target object. The detection assembly is configured to receive reflected light information passing through the receiving lens and reflected by the target object . The controller is electrically connected to the detection assembly, and the controller is configured to determine a measurement distance of the target object according to the reflected light information received by the detection assembly.