Detection of anomalous trailer behavior
The technology relates to determining whether a vehicle operating in an autonomous driving mode is experiencing an anomalous condition, for instance due to a loss of tire pressure, a mechanical failure, or a shift or loss of cargo. The actual current pose of the vehicle is compared to an expected pose of the vehicle, where the expected pose is based on a model of the vehicle. If a pose discrepancy is identified, the anomalous condition is determined from information associated with the pose discrepancy. The vehicle is then able to take corrective action based on the nature of the anomalous condition. The corrective action may include making a real-time driving change, modifying a planned route, alerting a remote operations center, or communicating with one or more other vehicles.
1 . A vehicle configured to operate in an autonomous driving mode, comprising:
memory including a vehicle model stored therein, the vehicle model being calibrated to account for at least one real-world condition or driving situation;
a driving system including a steering subsystem, an acceleration subsystem and a deceleration subsystem to control driving of the vehicle in the autonomous driving mode;
a perception system including one or more sensors, each sensor being positioned along the vehicle and configured to detect objects in an environment around the vehicle, the one or more sensors including at least one of a lidar sensor, a camera sensor or a radar sensor; and
a control system operatively connected to the driving system and the perception system, the control system having one or more computer processors configured to:
receive sensor data from the one or more sensors of the perception system while operating in the autonomous driving mode;
identify an actual pose of the vehicle along a roadway at a particular point in time based upon the received sensor data;
determine, using the vehicle model, a predicted pose of the vehicle at the particular point in time;
determine that there is an anomalous condition of the vehicle upon identification of a discrepancy between the actual pose of the vehicle along the roadway at the particular point in time and the predicted pose of the vehicle along the roadway at the particular point in time; and
effect a corrective action in response to the anomalous condition.
2 . The vehicle of claim 1 , wherein identification of the discrepancy includes a determination that there is a deviation from the predicted pose beyond a threshold deviation amount.
3 . The vehicle of claim 2 , wherein determination of the deviation from the predicted pose beyond the threshold deviation amount includes an evaluation of whether the deviation from the predicted pose persists over a given time interval.
4 . The vehicle of claim 1 , wherein identification of the discrepancy includes an evaluation of an amount of the discrepancy based on a current environmental condition.
5 . The vehicle of claim 1 , wherein identification of the discrepancy includes an evaluation of an accuracy of the actual pose.
6 . The vehicle of claim 1 , wherein identification of the discrepancy includes an evaluation of a cost factor associated with the corrective action.
7 . The vehicle of claim 6 , wherein the corrective action is deferred based on the evaluation until the vehicle reaches a specific location.
8 . The vehicle of claim 1 , wherein identification of the discrepancy includes an evaluation of whether a road condition is transient.
9 . The vehicle of claim 1 , wherein the vehicle model is a kinematic model.
10 . The vehicle of claim 1 , wherein the vehicle model includes a set of heuristics, the set of heuristics being associated with a history of other vehicles that operated along the roadway.
11 . The vehicle of claim 1 , wherein the vehicle model is a machine learning model, the machine learning model being trained to identify one or more types of anomalous conditions.
12 . The vehicle of claim 1 , wherein the anomalous condition is either a shift in cargo or a loss of cargo.
13 . The vehicle of claim 1 , wherein the anomalous condition is either a flat tire condition or a component mechanical failure.
14 . The vehicle of claim 1 , wherein the control system is configured to effect the corrective action in accordance with a type of cargo.
15 . The vehicle of claim 1 , further comprising:
a trailer; and
tractor including a coupling system configured to adjustably couple to the trailer.
16 . A method of operating a vehicle in an autonomous driving mode, the method comprising:
receiving, by one or more processors of a control system of the vehicle, sensor data from a perception system while operating in the autonomous driving mode;
identifying, by the one or more processors, an actual pose of the vehicle along a roadway at a particular point in time based upon the received sensor data;
determining, by the one or more processors using a vehicle model stored in memory, a predicted pose of the vehicle at the particular point in time, the vehicle model being calibrated to account for at least one real-world condition or driving situation;
determining, by the one or more processors, that there is an anomalous condition of the vehicle upon identifying a discrepancy between the actual pose of the vehicle along the roadway at the particular point in time and the predicted pose of the vehicle along the roadway at the particular point in time; and
effecting, by the one or more processors, a corrective action in response to the anomalous condition.
17 . The method of claim 16 , wherein the corrective action is a real-time corrective driving action.
18 . The method of claim 16 , wherein the corrective action includes at least one of modifying a planned route for the vehicle, alerting a remote operations center to the anomalous condition, or communicating with another vehicle regarding the anomalous condition.
19 . The method of claim 16 , wherein identifying the discrepancy includes a determination that there is a deviation from the predicted pose beyond a threshold deviation amount.
20 . The method of claim 16 , wherein identifying the discrepancy includes at least one of:
evaluating an amount of the discrepancy based on a current environmental condition;
evaluating an accuracy of the actual pose; or
evaluating a cost factor associated with the corrective action.
21 . The vehicle of claim 15 , wherein the at least one real-world condition or driving situation is associated with a positional relationship of the tractor and the trailer.
22 . The method of claim 16 , wherein the at least one real-world condition or driving situation is associated with a positional relationship of a tractor and a trailer of the vehicle.