# US 12,005,928 B2 · Dangerous road user detection and response > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,005,928 B2 - Document ID: US12005928B2 - Status: Granted - Kind: B2 - Application number: 17/398,528 - Filed: 2021-08-10 - Published: 2023-02-16 - Granted: 2024-06-11 - Earliest claimed: 2021-08-10 - Assignees: GM Cruise Holdings LLC - Applicants: GM Cruise Holdings LLC - Inventors: Travis McCawley; Christopher Ward; Saranya Konala; Nikola Noxon - Web page: https://patentgrep.com/patent/US12005928B2 ## CPC - B60K 35/22: Display screens - B60W 60/0015: {specially adapted for safety} - B60K 35/10: Input arrangements, i.e. from user to vehicle, associated with vehicle functions or specially adapted therefor - B60K 35/28: characterised by the type of the output information, e.g. video entertainment or vehicle dynamics information; characterised by the purpose of the output information, e.g. for attracting the attention of the driver - B60W 40/04: Traffic conditions - B60W 50/14: Means for informing the driver, warning the driver or prompting a driver intervention - B60W 60/0027: {using trajectory prediction for other traffic participants} - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/047: Probabilistic or stochastic networks - G06N 3/0475: Generative networks - G06N 3/084: Backpropagation, e.g. using gradient descent - G06N 3/094: Adversarial learning - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/00: Machine learning - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - B60K 2360/11: Instrument graphical user interfaces or menu aspects - B60K 2360/175: Autonomous driving - B60K 2360/178: Warnings - B60W 2050/146: {Display means} - B60W 2554/4046: Behavior, e.g. aggressive or erratic - B60K: ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES; ARRANGEMENT OR MOUNTING OF PLURAL DIVERSE PRIME-MOVERS IN VEHICLES; AUXILIARY DRIVES FOR VEHICLES; INSTRUMENTATION OR DASHBOARDS FOR VEHICLES; ARRANGEMENTS IN CONNECTION WITH COOLING, AIR INTAKE, GAS EXHAUST OR FUEL SUPPLY OF PROPULSION UNITS IN VEHICLES - B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Methods and systems are provided for detecting and responding to dangerous road users. In some aspects, a process can include steps for receiving sensor data of a detected object from an autonomous vehicle, determining whether the detected object is exhibiting a dangerous attribute, generating output data based on the determining of whether the detected object is exhibiting the dangerous attribute, and updating a machine learning model based on the output data relating to the dangerous attribute. ## Claims ### Claim 1 A computer-implemented method for detecting and responding to dangerous road users, the computer-implemented method comprising: receiving, at an autonomous vehicle system, sensor data of a detected object from an autonomous vehicle; generating, by the autonomous vehicle system, a three-dimensional (3D) bounding box around the detected object, wherein the 3D bounding box includes attributes, which comprise a dimension, a position, an orientation, a classification of the detected object, and a confidence of the 3D bounding box; determining, by the autonomous vehicle system, whether the detected object is exhibiting a dangerous attribute based on a behavior of the detected object and the attributes of the 3D bounding box; generating, by the autonomous vehicle system, output data based on the determining of whether the detected object is exhibiting the dangerous attribute; and updating, by the autonomous vehicle system, a machine learning model based on the output data relating to the dangerous attribute. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the detected object includes at least one of a motor vehicle, a cyclist, and a pedestrian. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the dangerous attribute includes at least one of swerving, crossing lane boundaries, and increased reaction time relative to a nominal reaction time. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, further comprising receiving a confirmation of the dangerous attribute of the detected object from a graphical user interface of the autonomous vehicle. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising receiving alert information based on the dangerous attribute of the detected object, the alert information including at least one of images, videos, and geographic positional data of the detected object. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, further comprising: receiving information relating to a person associated with the detected object; and updating the output data based on the information relating to the person associated with the detected object. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, further comprising providing the updated machine learning model to the autonomous vehicle to assist in automated evasive driving. ### Claim 8 A system for detecting and responding to dangerous road users, the system comprising: one or more processors; and at least one non-transitory computer-readable storage medium having stored therein instructions which, when executed by the one or more processors, cause the system to: receive sensor data of a detected object from an autonomous vehicle; generate a three-dimensional (3D) bounding box around the detected object, wherein the 3D bounding box includes attributes, which comprise a dimension, a position, an orientation, a classification of the detected object, and a confidence of the 3D bounding box; determine whether the detected object is exhibiting a dangerous attribute based on a behavior of the detected object and the attributes of the 3D bounding box; generate output data based on the determining of whether the detected object is exhibiting the dangerous attribute; and update a machine learning model based on the output data relating to the dangerous attribute. ### Claim 9 (depends on 8) The system of claim 8, wherein the detected object includes at least one of a motor vehicle, a cyclist, and a pedestrian. ### Claim 10 (depends on 8) The system of claim 8, wherein the dangerous attribute includes at least one of swerving, crossing lane boundaries, and increased reaction time relative to a nominal reaction time. ### Claim 11 (depends on 8) The system of claim 8, wherein the instructions which, when executed by the one or more processors, cause the system to receive a confirmation of the dangerous attribute of the detected object from a graphical user interface of the autonomous vehicle. ### Claim 12 (depends on 8) The system of claim 8, wherein the instructions which, when executed by the one or more processors, cause the system to receive alert information based on the dangerous attribute of the detected object, the alert information including at least one of images, videos, and geographic positional data of the detected object. ### Claim 13 (depends on 8) The system of claim 8, wherein the instructions which, when executed by the one or more processors, cause the system to: receive information relating to a person associated with the detected object; and update the output data based on the information relating to the person associated with the detected object. ### Claim 14 (depends on 8) The system of claim 8, wherein the instructions which, when executed by the one or more processors, cause the system to provide the updated machine learning model to the autonomous vehicle to assist in automated evasive driving. ### Claim 15 A non-transitory computer-readable storage medium comprising: instructions stored on the non-transitory computer-readable storage medium, the instructions, when executed by one more processors, cause the one or more processors to: receive sensor data of a detected object from an autonomous vehicle; generate a three-dimensional (3D) bounding box around the detected object, wherein the 3D bounding box includes attributes, which comprise a dimension, a position, an orientation, a classification of the detected object, and a confidence of the 3D bounding box; determine whether the detected object is exhibiting a dangerous attribute based on a behavior of the detected object and the attributes of the 3D bounding box; generate output data based on the determining of whether the detected object is exhibiting the dangerous attribute; and update a machine learning model based on the output data relating to the dangerous attribute. ### Claim 16 (depends on 15) The non-transitory computer-readable storage medium of claim 15, wherein the dangerous attribute includes at least one of swerving, crossing lane boundaries, and increased reaction time relative to a nominal reaction time. ### Claim 17 (depends on 15) The non-transitory computer-readable storage medium of claim 15, wherein the instructions, when executed by the one more processors, cause the one or more processors to receive a confirmation of the dangerous attribute of the detected object from a graphical user interface of the autonomous vehicle. ### Claim 18 (depends on 15) The non-transitory computer-readable storage medium of claim 15, wherein the instructions, when executed by the one more processors, cause the one or more processors to receive alert information based on the dangerous attribute of the detected object, the alert information including at least one of images, videos, and geographic positional data of the detected object. ### Claim 19 (depends on 15) The non-transitory computer-readable storage medium of claim 15, wherein the instructions, when executed by the one more processors, cause the one or more processors to: receive information relating to a person associated with the detected object; and update the output data based on the information relating to the person associated with the detected object. ### Claim 20 (depends on 15) The non-transitory computer-readable storage medium of claim 15, wherein the instructions, when executed by the one more processors, cause the one or more processors to provide the updated machine learning model to the autonomous vehicle to assist in automated evasive driving.