System and method to classify and remove object artifacts from light detection and ranging point cloud for enhanced detections
An enhanced light detection and ranging (LiDAR) assisted vehicle navigation system includes a global positioning system (GPS). A LiDAR device is in communication with the GPS generating and transmitting LiDAR signals reflected off a target proximate to a vehicle. Multiple vehicle sensors including at least a LiDAR sensor receive the LiDAR signals reflected off the target as a point cloud of data. A line profile of the target is used to identify if blooming is present. A characterization device performs a characterization analysis to identify an existence and extent of blooming of the LiDAR signals using the line profile of the target. A filter receives an output of the characterization device and removes the blooming if present to provide edge detection of the target.
1 . An enhanced light detection and ranging (LiDAR) assisted vehicle navigation system, comprising:
a global positioning system (GPS);
a LiDAR device in communication with the GPS generating and transmitting LiDAR signals reflected off a target proximate to a vehicle;
multiple vehicle sensors including at least a LiDAR sensor receiving the LiDAR signals reflected off the target as a point cloud of data;
a line profile of the target used to identify if blooming is present;
a characterization device performing a characterization analysis to identify an existence and extent of blooming of the LiDAR signals using the line profile of the target, wherein the characterization device includes the following:
a 3D plane-fit analyzer acquiring 3 dimensions of a size of the target and generating a 3D image of an area in a path of the LiDAR signals proximate to the vehicle;
a data serializer in communication with the 3D plane-fit analyzer converting a data object having a combination of code and data into a series of bytes to save a state of the target in a serialized form;
a horizontal intensity and reflectivity identification determiner receiving the state of the target in the serialized form and a vertical intensity and reflectivity identification determiner operating in parallel with the horizontal intensity and reflectivity identification determiner to identify an intensity and reflectivity of the target;
a signal to noise ratio (SNR) calculator identifying an SNR value of the point cloud of data using an output of the vertical intensity and reflectivity identification determiner and the horizontal intensity and reflectivity identification determiner;
a Z-score variation evaluator receiving an output of the SNR calculator and determining a Z-score, wherein the Z-score is defined relative to a mean;
a peak-to-valley deviation monitor receiving an output from the Z-score variation evaluator and measuring a difference between an ideal optic surface and an actual optic surface;
a decision device receiving the output of the peak-to-valley deviation monitor and determining that blooming is identified;
a coordinate identification unit identifying coordinates of the target when the blooming is identified; and
a filter receiving an output of the characterization device and removing the blooming to provide edge detection of the target.
2 . The enhanced light detection and ranging (LiDAR) assisted vehicle navigation system of claim 1 , further including a point cloud restoration device enhancing blooming filter data and aligning multiple objects to remove noise from the point cloud to define a clean target.
3 . The enhanced light detection and ranging (LiDAR) assisted vehicle navigation system of claim 2 , including a perception device receiving signal data from the point cloud restoration device and comparing the signal data to the clean target, and identifying instructions to be forwarded to the GPS system and to a control system of the vehicle.
4 . The enhanced light detection and ranging (LiDAR) assisted vehicle navigation system of claim 1 , including a measurement of 3 dimensions of a target size performed using the line profile of the target to identify a height, a width and a depth of the target.
5 . The enhanced light detection and ranging (LiDAR) assisted vehicle navigation system of claim 1 , further including a most likely match between the target and one or more known objects retrieved from a database saved in a cloud or from a memory analyzed by the characterization device.
6 . The enhanced light detection and ranging (LiDAR) assisted vehicle navigation system of claim 1 , wherein a point cloud noise at an edge of the target is removed by the filter, the filter further decreasing a latency to recognize the target and further processing the data of the point cloud to remove point cloud noise.
7 . A method to perform enhanced light detection and ranging (LiDAR) assisted vehicle navigation via a global positioning system (GPS) of a vehicle, comprising:
operating a LiDAR device in communication with the GPS to generate and transmit LiDAR signals to be reflected off a target proximate to the vehicle;
providing multiple vehicle sensors including at least a LiDAR sensor operating to receive the LiDAR signals reflected off the target as a point cloud of data;
operating a characterization device to perform a characterization analysis using a line profile of the target to identify an existence and an extent of data blooming of the LiDAR signals, wherein the characterization device includes the following:
a 3D plane-fit analyzer acquiring 3 dimensions of a size of the target and generating a 3D image of an area in a path of the LiDAR signals proximate to the vehicle;
a data serializer in communication with the 3D plane-fit analyzer converting a data object having a combination of code and data into a series of bytes to save a state of the target in a serialized form;
a horizontal intensity and reflectivity identification determiner receiving the state of the target in the serialized form and a vertical intensity and reflectivity identification determiner operating in parallel with the horizontal intensity and reflectivity identification determiner to identify an intensity and reflectivity of the target;
a signal to noise ratio (SNR) calculator identifying an SNR value of the point cloud of data using an output of the vertical intensity and reflectivity identification determiner and the horizontal intensity and reflectivity identification determiner;
a Z-score variation evaluator receiving an output of the SNR calculator and determining a Z-score, wherein the Z-score is defined relative to a mean;
a peak-to-valley deviation monitor receiving an output from the Z-score variation evaluator and measuring a difference between an ideal optic surface and an actual optic surface;
a decision device receiving the output of the peak-to-valley deviation monitor and determining that blooming is identified;
a coordinate identification unit identifying coordinates of the target when the blooming is identified; and
forwarding an output of the characterization device to a filter to remove the data blooming and to provide edge detection of the target.
8 . The method of claim 7 , further including:
enhancing blooming filter data and aligning multiple objects to remove noise from the point cloud of data to define a clean target using a point cloud restoration device; and
comparing the target to the clean target using a perception device receiving signal data from the point cloud restoration device.
9 . The method of claim 7 , further including following identification of the coordinates performing noise reduction by removing points from the target data to reduce noise.
10 . The method of claim 9 , further including following the noise reduction, forwarding noise reduction data to the point cloud of data to update the point cloud of data.
11 . The method of claim 7 , further including removing a region of spillover granular light proximate to the target.
12 . The method of claim 7 , further including applying the line profile to provide object recognition and classification of objects different from the target.