The present disclosure discloses a system and a method for controlling a device using a compound probabilistic filter. The method comprises collecting a sequence of measurements indicative of the state of the device at different control steps. Further, the method comprises executing iteratively a compound probabilistic filter configured to track the state of the device at each of the different control steps using the sequence of measurements to produce a sequence of states of the device corresponding to the sequence of measurements. Furthermore, the method comprises controlling the device using the tracked state of the device.

A receding horizon state estimator estimates state of a vehicle such as to reduce total communication cost of acquiring external measurements over a prediction horizon, in which state estimation accuracy for a time step is a function of state estimation accuracy for a previous time step. For each time step of the prediction horizon, estimator selects a subset of external sensors with external measurements sufficient to estimate the state with accuracy satisfying the constraint on state estimation accuracy for the corresponding time step while reducing a total communication cost of acquiring the external measurements over the prediction horizon. The estimator requests the external measurements from the subset of external sensors determined for a current time step and estimates the state of the vehicle using the internal and the requested external measurements.

Systems and methods are provided for determining a lane assignment for an autonomous vehicle along a road segment. In one implementation, at least one image representative of an environment of the vehicle is received from a camera. The at least one image may be analyzed to identify at least one recognized landmark, and an indicator of a lateral offset distance between the vehicle and the at least one recognized landmark may be determined. Moreover, a lane assignment of the vehicle along the road segment may be determined based on the indicator of the lateral offset distance between the vehicle and the at least one recognized landmark.

The embodiments described herein aim to improve environmental sensing by providing a computationally efficient and accurate means for fusing sensor data and using this fused data to control sensors to focus on areas that would most reduce the uncertainty in the sensing system. In this way, the system can direct sensors to focus on the most important areas and features within the environment in order to provide the most effective sensor data (e.g. for use by a control system). The methods described herein make use of multi-agent sensor-action fusion. The methods are multi-agent in that a set of machine learning agents are trained in order to control the sensors to focus on the most important features and regions. The embodiments implement sensor-action fusion in that sensor fusion is performed in order to obtain a combined view of the environment and this combined view is utilised to determine the most appropriate actions.

Systems and methods are provided for constructing, using, and updating the sparse map for autonomous vehicle navigation. In one implementation, a non-transitory computer-readable medium includes a sparse map for autonomous vehicle navigation along a road segment. The sparse map includes a polynomial representation of a target trajectory for the autonomous vehicle along the road segment and a plurality of predetermined landmarks associated with the road segment, wherein the plurality of predetermined landmarks are spaced apart by at least 50 meters. The sparse map has a data density of no more than 1 megabyte per kilometer.

This application discloses a computing system to implement sensor event detection and fusion system in an assisted or automated driving system of a vehicle. The computing system can monitor an environmental model to identify spatial locations in the environmental model populated with temporally-aligned measurement data. The computing system can analyze, on a per-sensor basis, the temporally-aligned measurement data at the spatial locations in the environmental model to detect one or more sensor measurement events. The computing system can utilize the sensor measurement events to identify at least one detection event indicative of an object proximate to the vehicle. The computing system can combine the detection event with at least one of another detection event, a sensor measurement event, or other measurement data to generate a fused detection event. A control system for the vehicle can control operation of the vehicle based, at least in part, on the detection event.

A low clearance detection system for a vehicle. In one example, the system includes a first sensor configured to detect an object in front of the vehicle and generate an object clearance signal. The system includes a second sensor configured to detect a load height of a load of the vehicle and generate a load height signal. The system includes a third sensor configured to detect a distance between the third sensor to a ground surface and to generate a ground reference signal. The system also includes an electronic processor configured to receive the object clearance signal, the load height signal, and the ground reference signal. The electronic processor determines an object clearance threshold, a clearance height of the load, and a load collision condition when the clearance height exceeds the object clearance threshold. In response to determining the load collision condition, the electronic processor controls a vehicle system.

A system controls a vehicle. The vehicle implements at least one control application using a variable measured by at least one sensor of a perception system installed in the vehicle. The control system includes an adaptive controller to dynamically activate one or more elementary controllers of a set of elementary controllers including at least two elementary controllers, each elementary controller can apply a control function of a vehicle parameter acting on actuators of the vehicle, the control functions being separate, based on a precision indicator of the perception system determined according to a real-time value of the variable.

An autonomous vehicle controlled based upon the output of a trained object classifier is described herein. The object classifier is trained using labeled training data generated by a pipeline configured to assign labels to unlabeled sensor data. The pipeline includes transmitting sensor signal data capturing an object to individual computing devices for indications of an object type, wherein a label is assigned to the object based on the indications and provided to a data store as labeled training data. A learning system receives the labeled training data and generates a trained object classifier (e.g., a neural network) that is deployed in an autonomous vehicle to control operation of a mechanical system based on an output thereof.

Detection of merge scenarios by an autonomous vehicle (AV) is disclosed. A system includes a memory and a processor. The memory includes instructions executable by the processor to, in response to detecting a merge scenario, identify an interacting object pair including a merging object and a crossing object, where the AV is the merging object; generate a yield hypothesis and a no-yield hypothesis; compute a yield reference path corresponding to the yield hypothesis and a no-yield reference path corresponding to the no-yield hypothesis; determine a yield likelihood of the yield hypothesis and a no-yield likelihood of the no-yield hypothesis; and operate the AV to merge or to wait until the merge scenario is no longer detected based on the yield likelihood and the no-yield likelihood.