Suteng Innovation Technology (RoboSense)

A chip module includes a packaging substrate, a system-on-chip disposed on a surface of the packaging substrate, a laser emission chip disposed on the system-on-chip and configured to emit laser signals, and at least two receiving chips disposed on the system-on-chip. The receiving chips include a first receiving chip configured to receive echo laser signals and a second receiving chip configured to receive visible light signals. A cover is disposed on the packaging substrate, wherein the cover and the packaging substrate jointly enclose to form an installation cavity, the system-on-chip, the laser emission chip, the first receiving chip, and the second receiving chip are all located within the installation cavity. The cover is provided with a first region for laser signals to pass through, a second region for echo laser signals to pass through, and a third region for visible light signals to pass through.

A single-photon avalanche diode array, a receiving sensor and a LiDAR are provided, wherein at least two SPAD units are arranged in an array, a micro lens converges the incident light onto the corresponding SPAD unit, a back metal grid connects the SPAD unit with the corresponding external electrode, a first dielectric layer is arranged between the micro lens and the SPAD unit, a second dielectric layer is arranged between the SPAD unit and a front metal wiring layer, and the front metal wiring layer is electrically connected to the corresponding SPAD unit through a contact metal wire. By arranging deep groove separation columns between adjacent SPAD units, thereby reducing the probability of photon crosstalk and reducing the optical crosstalk of the device.

The present application provides an echo signal processing method. The method includes: acquiring single histograms obtained from at least two scans; acquiring a superimposed histogram according to the single histograms; respectively extracting echo information of the superimposed histogram and echo information of target single histograms; calculating an echo similarity of each echo in the superimposed histogram and an echo similarity of corresponding echoes between the target single histograms according to the echo information of the superimposed histogram and the echo information of the target single histograms; determining an echo confidence level of each echo in the superimposed histogram according to the echo similarity of the echoes in the superimposed histogram and the echo similarity between the target single histograms; and identifying a valid echo according to the echo confidence level of the echo. This improves the identification accuracy of a radar system for noise signals and crosstalk echoes.

An emission module (1) includes a carrier board (11) and a light-emitting module (12) having a first subgroup (12A) and a second subgroup (12B). The first subgroup comprises at least two first emission units (121), at least one first drive unit (122), and at least one first energy storage unit (123). The first drive unit drives the at least two first emission units to emit laser light using energy stored in the first energy storage unit. The second subgroup comprises at least two second emission units (124), at least one second drive unit (125), and at least one second energy storage unit (126). The second drive unit drives the at least two second emission units to emit laser light using energy stored in the second energy storage unit. The first emission units and the second emission units are arranged in a staggered manner along a second direction. A single drive unit concurrently drives multiple emission units, compressing the light-emitting module's volume.

A LiDAR and an autonomous driving device are provided. The LiDAR includes a transceiver module and a rotating mirror. The transceiver module includes a transmitting unit, a receiving unit, a first housing, a second housing, and a plane mirror. The first housing includes a receiving cylinder and a transmitting cylinder. The second housing includes a first sub-housing and a second sub-housing. The transmitting cylinder and the first sub-housing form a transmitting channel, and the receiving cylinder and the second sub-housing form a receiving channel. The first sub-housing includes a first opening. The second sub-housing includes a second opening and a first boss. The plane mirror is fixed to the first opening, and the first boss extends from a first end of the second opening to a second end of the second opening.

A distance and reflectivity measurement method, apparatus, storage medium, and lidar are disclosed. The lidar acquires an echo signal from light reflected by a target object. A distance threshold measures the echo signal to determine the target's distance relative to the lidar. A reflectivity threshold, higher than the distance threshold, measures the echo signal to determine the target's reflectivity. Using separate thresholds, with the reflectivity threshold being higher, provides high accuracy for distance measurement and high differentiation for reflectivity measurement, improving the overall accuracy of calculating both distance and reflectivity for the target object.

An azimuth angle calibration method for a LiDAR, an azimuth angle calibration apparatus, and a LiDAR are disclosed. The method includes acquiring calibration point cloud data comprising at least one frame of point cloud data from the LiDAR. A measured azimuth angle is determined based on the calibration point cloud data, where the measured azimuth angle corresponds to a calibration zone. A compensation coefficient is calculated using the measured azimuth angle and a preset angle value. Angle compensation is then applied to azimuth angle information of detection point cloud data based on the compensation coefficient, wherein the detection point cloud data is generated during LiDAR operation. This method improves control accuracy in LiDAR scanning processes by compensating for directional angle errors.

A distance measurement method, applied to a data processing module coupled to a receiving module, includes: determining a plurality of received data based on an echo signal received by the receiving module, wherein the echo signal is reflected by a target object after a transmitting module emits a laser pulse toward the target object, and the received data are derived from an electrical signal output by the receiving module in response to the echo signal; organizing the received data into N data groups, each group comprising at least one received data, with at least two groups sharing partially identical received data, where N is an integer greater than or equal to 2; and computing a distance measurement result for each data group based on its respective received data. This method enhances distance measurement capability without requiring changes to radar hardware configuration.

A photoelectric conversion device, a fabrication method, and an image sensor are disclosed. The device includes a substrate with at least two avalanche diode units. Each unit has a device region surrounded by a back-side deep trench isolation structure. At least one front-side trench isolation structure is disposed between any two adjacent units. A doped region, formed by outward diffusion from the front-side trench isolation structure, has a gradually decreasing doping concentration gradient. At least a portion of the doped region extends beyond the back-side deep trench isolation structure to form a dark current suppression region within each adjacent avalanche diode unit. This configuration reduces the dark count rate (DCR) of the device, thereby enhancing its electrical performance and reliability.

The present application discloses an abnormal field of view recognition method and device, a storage medium and a MEMS LiDAR device, wherein the method comprises: respectively adjusting angles of a galvanometer with respect to an X axis and a Y axis when the MEMS LiDAR device is started, acquiring echo data from scanning a window at the adjusted angles of the galvanometer with respect to the X axis and the Y axis; acquiring, from the echo data, distances between the MEMS LiDAR device and respective ones of an upper edge, a lower edge, a left edge and a right edge of the window; and determining whether a field of view of the galvanometer is abnormal based on the acquired distances. According to the present application, abnormality of the field of view of the galvanometer can be detected when the MEMS LiDAR device is started, so that the security problem caused by the abnormal field of view of the galvanometer is avoided, and safety and reliability of the MEMS LiDAR device are improved.