Low-impact collision detection

A device configured to detect low-impact collisions, the device comprising:

• means for determining whether an object collided with a vehicle based on a determination of a first value of a motion parameter based on data received from at least one motion sensor configured to measure at least an acceleration of the vehicle, a determination of a second value of the motion parameter based on data received from a plurality of level sensors, wherein each level sensor is configured to measure a relative position between a body of the vehicle and a respective wheel of a plurality of wheels of the vehicle, and an application of a machine-trained model to the first value of the motion parameter and the second value of the motion parameter, wherein the application of the machine-trained model comprises a comparison of the first value of the motion parameter with the second value of the motion parameter and a determination whether the comparison indicates a collision or a non-collision, wherein the motion parameter includes one or more of:
• a roll rate of the vehicle, a pitch angle or pitch rate of the vehicle, a heave position of the vehicle, or an acceleration rate of the vehicle;
• and
• means for performing one or more actions in response to determining the object collided with the vehicle.

A method comprising:

• determining whether an object collided with a vehicle based on an application of a machine-trained model to a first measurement of a motion parameter and a second measurement of the motion parameter,
• wherein the application of the machine-trained model comprises a comparison of the first measurement of the motion parameter with the second measurement of the motion parameter and a determination whether the comparison indicates a collision or a non-collision,
• wherein the motion parameter includes one or more of:
• a roll rate of the vehicle, a pitch angle or pitch rate of the vehicle, a heave position of the vehicle, or an acceleration rate of the vehicle,
• wherein the first measurement of the motion parameter is determined based on data received from at least one motion sensor configured to measure at least an acceleration of the vehicle,
• wherein the second measurement of the motion parameter is determined based on data received from a plurality of level sensors, wherein each level sensor is configured to measure a relative position between a body of the vehicle and a respective wheel of a plurality of wheels of the vehicle;
• and
• performing one or more actions in response to determining that the object collided with the vehicle.

A non-transitory computer-readable storage medium comprising instructions that, when executed by at least one processor of a computing system of a vehicle, cause the at least one processor to:

• determine whether an object collided with a vehicle based on an application of a machine-trained model to a first measurement of a motion parameter and a second measurement of the motion parameter, the application comprising a comparison of the first measurement of the motion parameter with the second measurement of the motion parameter and a determination whether the comparison indicates a collision or a non-collision, wherein the motion parameter includes one or more of:
• a roll rate of the vehicle, a pitch angle or pitch rate of the vehicle, a heave position of the vehicle, or an acceleration rate of the vehicle, wherein the first measurement of the motion parameter is determined based on data received from at least one motion sensor configured to measure at least an acceleration of the vehicle, wherein the second measurement of the motion parameter is determined based on data received from a plurality of level sensors, wherein each level sensor is configured to measure a relative position between a body of the vehicle and a respective wheel of a plurality of wheels of the vehicle;
• and
• perform one or more actions in response to determining the object collided with the vehicle.