Systems and techniques for vehicle inspection and condition analysis
Systems and techniques for inspecting vehicles, such as cars or trucks. Some embodiments provide a method for inspecting a vehicle using a vehicle inspection system. The vehicle inspection system may include a vehicle exterior inspection system having a plurality of exterior sensor arrays, the plurality of exterior sensor arrays including respective sets of sensors oriented in different directions. The vehicle inspection system may include a vehicle undercarriage inspection system having an undercarriage sensor array including cameras configured to capture images at different angles, and a sensor configured to output vehicle position/motion data. In some embodiments, the method includes: moving the vehicle and/or vehicle inspection system; triggering capture, by the exterior sensor arrays and undercarriage sensor array, of sensor data about the vehicle; capturing the sensor data; stopping capture of the sensor data; stopping movement of the vehicle and/or vehicle inspection system; and generating a vehicle condition report.
1 . A system for vehicle inspection, the system comprising:
a vehicle exterior inspection system, comprising:
a first support member, comprising:
a first set of cameras configured to image an exterior of a vehicle, the first set of cameras comprising:
a first camera oriented to capture images of the vehicle as the vehicle approaches the first support member;
a second camera oriented to capture images of the vehicle as the vehicle is alongside the first support member; and
a third camera oriented to capture images of the vehicle as the vehicle moves past the first support member;
a second support member, comprising:
a second set of cameras configured to image at least one wheel and/or tire of the vehicle; and
a wheel light source configured to illuminate the at least one wheel and/or tire of the vehicle; and
an undercarriage imaging device, comprising:
a third set of cameras configured to image an undercarriage of the vehicle, the third set of cameras including a first, second, and third camera oriented to capture images at different angles relative to the undercarriage of the vehicle; and
an undercarriage light source configured to illuminate the undercarriage of the vehicle; and
a processor configured to:
receive image data from the first set of cameras, the second set of cameras, and the third set of cameras; and
process the image data using at least one trained machine learning model to determine one or more defects of the vehicle.
2 . The system of claim 1 , wherein the at least one trained machine learning model comprises:
a vehicle exterior machine learning model trained on images of vehicle exteriors;
a wheel and tire machine learning model trained on images of wheels and/or tires; and
an undercarriage machine learning model trained on images of vehicle undercarriages.
3 . The system of claim 1 , wherein the first support member further comprises a vehicle exterior light source configured to illuminate the exterior of the vehicle, and/or wherein the first support member further comprises an infrared camera.
4 . The system of claim 1 , wherein the processor is configured to perform:
identifying, from the image data, an image containing personally identifiable information (PII);
identifying a region of the image containing the PII; and
distorting the region of the image containing the PII.
5 . The system of claim 1 , wherein;
the first support member further comprises a time of flight sensor, and
the processor is configured to trigger capture of the image data based on an output of the time of flight sensor.
6 . The system of claim 1 , wherein the one or more defects include: scratches to an exterior of the vehicle, cracked windows, mirrors, or windshields, chipped paint, dents to the exterior of the vehicle, misaligned body panels, missing vehicle parts, non-standard replacement parts, non-standard paint, aftermarket vehicle accessories, rust/corrosion on the vehicle, damaged wheels, damaged tires, bald tires, tire sidewall bubbles, broken tire valves, wheel misalignment, mismatched tires, brake rotor discoloration, brake rotor damage, brake rotor wear, and/or suspension modifications.
7 . The system of claim 1 , wherein the processor is further configured to:
generate, using photogrammetry, neural radiance fields or Gaussian splatting, a 3D model of at least a portion of the vehicle using the image data; and
generate a visualization of the 3D model.
8 . The system of claim 1 , wherein the image data comprises a plurality of images, wherein the processor is further configured to perform:
identifying a subset of the plurality of images for subsequent processing to identify whether the vehicle has one or more defects, wherein the identifying is performed based on a pose of the vehicle in images of the plurality of images, the identifying comprising:
determining a degree of matching between the pose of the vehicle in images of the plurality of images and vehicle poses of a reference set of vehicle poses; and for each vehicle pose of the reference set of vehicle poses, identifying an image in the plurality of images having at least a threshold degree of matching and including the identified image in the subset of the plurality of images; and
processing the subset of the plurality of images using at least one trained machine learning model to determine one or more defects of the vehicle.
9 . A system for vehicle inspection, the system comprising:
a vehicle exterior inspection system, comprising:
a first support member, comprising:
a first set of cameras configured to image an exterior of a vehicle;
a second support member, comprising:
a second set of cameras configured to image at least one wheel and/or tire of the vehicle; and
a wheel light source configured to illuminate the at least one wheel and/or tire of the vehicle; and
a set of microphones configured to capture audio of the vehicle; and
a processor configured to:
receive image data from the first set of cameras and the second set of cameras;
receive audio data from the set of microphones; and
process at the image data and the audio data using at least one trained machine learning model to determine one or more defects of the vehicle.
10 . The system of claim 9 , wherein the set of microphones comprises a plurality of microphones configured to capture audio from different portions of the vehicle.
11 . The system of claim 9 , wherein the at least one trained machine learning model comprises:
a vehicle exterior machine learning model trained on images of vehicle exteriors; and
a wheel and tire machine learning model trained on images of wheels and/or tires.
12 . The system of claim 9 , wherein the at least one trained machine learning model is configured to output:
the one or more defects of the vehicle; and
one or more likelihoods that the one or more defects of the vehicle are present.
13 . The system of claim 9 , wherein the vehicle exterior inspection system comprises an undercarriage imaging device, comprising:
a third set of cameras configured to image an undercarriage of the vehicle, the third set of cameras including a first, second, and third camera oriented to capture images at different angles relative to the undercarriage of the vehicle; and
an undercarriage light source configured to illuminate the undercarriage of the vehicle,
wherein the processor is further configured to receive image data from the third set of cameras.
14 . The system of claim 13 , wherein the processor is configured to generate a composite image of the undercarriage of the vehicle by combining images captured by the third set of cameras based on speed of the vehicle.
15 . The system of claim 13 , wherein the at least one trained machine learning model comprises an undercarriage machine learning model trained on images of vehicle undercarriages.
16 . A system for vehicle inspection, the system comprising:
a vehicle exterior inspection system, comprising:
a first support member, comprising:
a first set of cameras configured to image at least one wheel and/or tire of a vehicle; and
a wheel light source configured to illuminate the at least one wheel and/or tire of the vehicle;
an undercarriage imaging device, comprising:
a second set of cameras configured to image an undercarriage of the vehicle, the second set of cameras including a first, second, and third camera oriented to capture images at different angles relative to the undercarriage of the vehicle;
an undercarriage light source configured to illuminate the undercarriage of the vehicle; and
at least one microphone; and
at least one sensor configured to collect alignment data of the vehicle relative to the vehicle exterior inspection system; and
a processor configured to:
receive alignment data from the at least one sensor;
trigger, based on the alignment data, capture of images by the first set of cameras and the second set of cameras;
receive image data from the first set of cameras and the second set of cameras;
process the image data using at least one trained machine learning model to determine one or more defects of the vehicle;
receive audio data from the at least one microphone; and
process the audio data using the at least one trained machine learning model to determine the one or more defects of the vehicle.
17 . The system of claim 16 , wherein the at least one sensor comprises a ranging sensor.
18 . The system of claim 16 , wherein the undercarriage light source comprises a plurality of individual light sources configured to illuminate at different angles relative to the undercarriage of the vehicle, wherein the undercarriage light source comprises constant current light emitting diodes.
19 . The system of claim 16 , wherein:
the vehicle exterior inspection system further comprises:
a second support member, comprising:
a third set of cameras configured to image an exterior of a vehicle, and the processor is further configured to receive image data from the third set of cameras.
20 . The system of claim 16 , wherein:
the at least one microphone is a plurality of microphones, and
the plurality of microphones are distributed along a width of the undercarriage imaging device.