Search

DE-112024002075-T5 - DETECTION AND ASSESSMENT OF LIDAR PERFORMANCE DECLINE

DE112024002075T5DE 112024002075 T5DE112024002075 T5DE 112024002075T5DE-112024002075-T5

Abstract

A LIDAR system comprising one or more light sources configured to project light in the direction of a field of view (FOV) of the LIDAR system, one or more sensors configured to receive light projected by the light source(s) and reflected by objects in the FOV, and one or more processors. The processor(s) is/are configured to cause the light source(s) to project light in a multitude of sampling cycles in the direction of the field of view, to receive reflection signals from the sensor(s) indicating at least a portion of the projected light reflected by objects in the field of view, to identify volumetrically dispersed targets indicative of at least one environmental condition in the FOV based on a statistical analysis of data derived from the reflection signals received during the multitude of sampling cycles, and to send one or more alert messages indicating the presence of the environmental conditions to one or more systems associated with a vehicle on which the LIDAR system is mounted.

Inventors

  • Ronen Eshel
  • Nir Goren

Assignees

  • INNOVIZ TECHNOLOGIES LTD.

Dates

Publication Date
20260513
Application Date
20240513
Priority Date
20230514

Claims (20)

  1. A LIDAR system comprising: at least one light source configured to project light onto a field of view of the LIDAR system; at least one sensor configured to receive light projected by the at least one light source and reflected by at least one object in the field of view; and at least one processor configured to: cause the at least one light source to project light onto at least one portion of the field of view in a plurality of sampling cycles; receive reflection signals from the at least one sensor indicative of at least a portion of the projected light reflected by at least one object in at least one portion of the field of view; Volumetrically dispersed targets are identified in at least one portion of the field of view based on a statistical analysis of data derived from the signals generated by the at least one sensor during the plurality of sampling cycles, wherein the volumetrically dispersed targets are indicative of at least one environmental condition; and at least one warning message is sent to at least one system associated with a vehicle on which the LIDAR system is mounted, wherein the at least one warning message indicates the presence of the at least one environmental condition.
  2. LiDAR system according to Claim 1 , where at least one environmental condition is an element of a group consisting of ice, snow, rain, hail, dust and fog.
  3. LIDAR system according to one of the preceding claims, wherein the statistical analysis indicates at least one characteristic of the at least one environmental condition, wherein the at least one characteristic is an element of a group consisting of: a precipitation density, a particle density and an average particle size.
  4. A LIDAR system according to one of the preceding claims, wherein the statistical analysis comprises determining a variation in an observed level of at least one detection parameter of the LIDAR system induced by at least one characteristic of at least one environmental condition, wherein the at least one detection parameter is an element from a group consisting of: a reflectivity level of at least one object identified in at least one section of the field of view, a detection area, an error detection rate, and a confidence level of the detection.
  5. LIDAR system according to one of the preceding claims, wherein the statistical analysis further comprises determining the variation in the observed level of the at least one detection parameter in combination with a distance between the LIDAR system and the at least one identified object in order to identify a dependency that specifies the at least one environmental condition.
  6. A LIDAR system according to one of the preceding claims, wherein the statistical analysis comprises determining at least one change in a noise baseline over a range of distances relative to the LIDAR system, wherein the at least one change in the noise baseline is indicative for at least one volumetric reflection condition induced by the at least one environmental condition.
  7. LIDAR system according to one of the preceding claims, wherein the statistical analysis comprises calculating at least one light reflection distribution pattern that indicates at least one volumetric reflection condition caused by the at least one environmental condition.
  8. A LIDAR system according to one of the preceding claims, further comprising applying a statistical analysis to analyze data extracted from a point cloud created on the basis of the reflection signals in order to identify at least one point which has no neighboring points and is therefore possibly indicative of the at least one environmental condition.
  9. A LIDAR system according to one of the preceding claims, further comprising estimating the magnitude of an impairment of at least one operational capability of the LIDAR system based on statistical analysis, wherein the at least one operational capability is an element of a group consisting of: a detection range and an accuracy of a certain distance to at least one object identified in at least one section of the field of view.
  10. LIDAR system according to one of the preceding claims, further comprising predictions, based on statistical analysis, of an expected drop in the operational capability of at least one below a predetermined performance threshold.
  11. LIDAR system according to one of the preceding claims, wherein the expected magnitude of the impairment of the at least one operational capability is predicted on the basis of an analysis of information derived from the data and compared with reference information retrieved from at least one lookup table.
  12. LIDAR system according to one of the preceding claims, wherein the expected magnitude of the impairment of the at least one operational capability is predicted using at least one machine learning model trained to estimate the magnitude of the impairment of the at least one operational capability, wherein the at least one machine learning model is trained using a training data set comprising light reflection distribution patterns indicative of light reflection through the volumetrically dispersed targets.
  13. LIDAR system according to one of the preceding claims, further comprising predictions, based on statistical analysis, of an expected drop in the operational capability of at least one below a predetermined performance threshold.
  14. A LIDAR system according to one of the preceding claims, which further comprises, on the basis of statistical analysis, predicting a time period until at least one operational capability is expected to fall below a predetermined performance threshold.
  15. A LIDAR system according to one of the preceding claims, further comprising the identification of the at least one environmental condition based on statistical analysis in combination with sensor data acquired by at least one external sensor associated with a vehicle on which the LIDAR system is mounted, wherein the at least one external sensor is an element of a group consisting of: an external light source, an ambient light sensor and a precipitation sensor.
  16. A LIDAR system according to one of the preceding claims, further comprising identifying the at least one environmental condition and/or an effect of the at least one environmental condition on the performance of the LIDAR system based on statistical analysis in combination with data associated with the location of a vehicle on which the LIDAR system is mounted, wherein the location of the vehicle is derived from at least one of the following: a navigation system of the vehicle and a map database, wherein the data associated with the location of the vehicle are received from at least one remote system.
  17. A LIDAR system according to one of the preceding claims, further comprising identifying, based on statistical analysis, the presence of at least one blocking agent on a window associated with the LIDAR system, wherein the blocking agent is an element of a group consisting of: ice, water droplets, smog, spray mist, dust, pollen, insects, mud and bird droppings.
  18. LIDAR system according to one of the preceding claims, further comprising predicting, based on statistical analysis, a time period until the at least one operational capability is expected to fall below the predetermined performance threshold, based on an accumulation rate of the at least one blocking agent on the window.
  19. LIDAR system according to one of the preceding claims, wherein the at least one processor is further configured to adapt the at least one warning to provide at least one recommended operating restriction for a vehicle to which the LIDAR system is attached.
  20. A method for detecting environmental conditions based on a statistical analysis of data acquired by a LiDAR system, comprising: causing at least one light source of a LiDAR system to project light onto at least one portion of a field of view of the LiDAR system in a plurality of scanning cycles; receiving reflection signals from at least one sensor of the LiDAR system, indicating at least a portion of the projected light reflected by at least one object in at least one portion of the field of view; identifying volumetrically dispersed targets in at least one portion of the field of view based on a statistical analysis of data derived from the received signals generated by the at least one sensor during the plurality of scanning cycles, wherein the volumetrically dispersed targets are indicative of at least one environmental condition; and Sending at least one warning message to at least one system that is connected to a vehicle on which the LIDAR system is mounted, wherein the at least one warning message is indicative of the presence of at least one environmental condition.

Description

TECHNICAL AREA The present disclosure relates to a technology for scanning a surrounding environment, in particular, but not exclusively, to the use of LIDAR-based systems for scanning an environment in order to detect objects in that surrounding environment. RELATED REGISTRATIONS The present application claims priority over the provisional application US patent application no. 63/502,100 , submitted on 14 May 2023, and the preliminary US patent application no. 63/503,479 , submitted on May 21, 2023, the contents of which are hereby incorporated in full by reference. TECHNICAL BACKGROUND With the advent of driver assistance systems and autonomous vehicles, automobiles must be equipped with systems capable of reliably perceiving and interpreting their surroundings, including the identification of obstacles, hazards, objects, and other physical parameters that could impair the vehicle's navigation. Various technologies have been proposed for this purpose, including, for example, Radio Detection and Ranging (RADAR), LiDAR, camera-based systems, and/or similar technologies, which can be operated individually, in combination, and/or redundantly. A major challenge for advanced driver assistance systems (ADAS) and autonomous vehicle systems (AV) is their ability to reliably, accurately and/or consistently determine the vehicle's surroundings under various environmental conditions, including, for example, rain, snow, ice, hail, fog, smog, dust, insects, darkness, bright light and/or the like. LIDAR technology can function well under such diverse conditions because it is based on mapping objects in the vehicle's environment and measuring distances to the detected objects by, for example, actively projecting laser light (continuous and/or pulsed) into the vehicle's surroundings and measuring the light reflected by objects in the environment. SUMMARY The objective of this disclosure is to provide methods, systems, and/or software program products for improving the performance of LiDAR systems by detecting and evaluating performance degradation, including degradation caused by adverse environmental conditions. This objective is achieved by the features of the independent claims. Further embodiments are apparent from the dependent claims, the description, and the figures. It should be noted that several such embodiments can be combined into a single embodiment. According to a first aspect of the embodiments disclosed herein, a LIDAR system is provided comprising: one or more light sources configured to project light onto a field of view of the LIDAR system; one or more sensors configured to receive light projected by the one or more light sources and reflected by one or more objects in the field of view; and one or more processors configured to: cause the one or more light sources to project light onto at least a portion of the field of view in a plurality of sampling cycles; receive reflection signals from the one or more sensors indicative of at least a portion of the projected light reflected by one or more objects in the at least one portion of the field of view; and identify volumetrically dispersed targets in at least one portion of the field of view based on a statistical analysis of data derived from the signals generated by the one or more sensors during the plurality of sampling cycles. The volumetrically dispersed targets indicate one or more environmental conditions, and transmit one or more alert messages to one or more systems associated with a vehicle on which the LIDAR system is mounted, the alert message indicating the presence of the one or more environmental conditions. According to a second aspect of the embodiments disclosed herein, a method for detecting environmental conditions is provided based on a statistical analysis of data acquired by a LIDAR system, the method comprising: causing one or more light sources of a LIDAR system to project light in a plurality of scanning cycles onto at least one portion of a field of view of the LIDAR system; receiving from one or more sensors of the LIDAR system reflection signals indicative of at least a portion of the projected light emitted by one or more objects in at least one area The system is designed to: identify volumetrically dispersed targets in at least one section of the field of view based on a statistical analysis of data derived from the received signals generated by the one or more sensors during the multitude of sampling cycles, wherein the volumetrically dispersed targets are indicative of one or more environmental conditions; and send one or more alert messages to one or more systems associated with a vehicle on which the lidar system is mounted. The one or more alert messages indicate the presence of the one or more environmental conditions. In a further embodiment of the first and/or second aspect, optionally together with one or more of its associated embodiments, the one or more environmental conditions are elements of a group consistin