Abstract: Upon detecting a mobile object approaching a target region, a target speed profile for a vehicle, a position and a speed for the vehicle, and a position and a speed for the mobile object are input to a first neural network that outputs an entry probability distribution of predicted entry times at which the mobile object will enter the target region. The target speed profile, the position and the speed for the vehicle, the position and the speed for the mobile object, and a predicted entry time at which the mobile object entered the target region are input to a second neural network that outputs an exit probability distribution of predicted exit times at which the mobile object will exit the target region. An optimized speed profile is determined based on the entry and exit probability distributions by executing a model predictive control algorithm in a rolling horizon. A vehicle is operated based on the optimized speed profile.

Abstract: A computer includes a processor and a memory, and the memory stores instructions executable by the processor to receive sensor data indicating an obstacle, formulate a first control barrier function for a vehicle based on the sensor data, formulate a second control barrier function for a trailer hitched to the vehicle based on the sensor data, determine a control input based on the first and second control barrier functions, and actuate a component of the vehicle according to the control input.

Abstract: A computer includes a processor and a memory, the memory storing instructions executable by the processor to receive sensor data indicating an obstacle, formulate a control barrier function for a vehicle based on the sensor data, determine a control input based on the control barrier function, and actuate a component of the vehicle according to the control input. The control barrier function is defined with respect to a reference point that is spaced from a centroid of the vehicle.

Abstract: A lateral virtual boundary for a host vehicle is identified based on a lateral distance between the host vehicle and a target vehicle, a longitudinal distance between the host vehicle and the target vehicle, and a speed of the target vehicle relative to the host vehicle. A forward virtual boundary for the host vehicle is identified based on the longitudinal distance between the host vehicle and the target vehicle. A lateral constraint value of the lateral virtual boundary and a forward constraint value of the forward virtual boundary are determined. A longitudinal acceleration and a steering angle are determined based on the lateral and forward virtual boundaries and the lateral and forward constraint values. One or both of a steering component or a brake are actuated based on the longitudinal acceleration and the steering angle.

Abstract: Upon detecting a lane-change trigger to move an ego vehicle from a current lane to a target lane, a target vehicle is identified in the target lane. Then a set of candidate trajectories is built for the lane-change maneuver by, for each of the target vehicles in the target lane: determining a set of constraints for a lane-change maneuver based on current data collected in the ego vehicle, including speed and distance constraints for the ego vehicle, and adding to the set of candidate trajectories, from a stored set of trajectories, a candidate trajectory for the lane-change maneuver that satisfies the constraints for the lane-change maneuver. Upon determining that the set of candidate trajectories has been built for each of the target vehicles, an optimal trajectory is selected from the set of candidate trajectories. A longitudinal acceleration of the ego vehicle is commanded based on the optimal trajectory.

Abstract: Upon detecting a mobile object approaching a target region, a target speed profile for a vehicle, a position and a speed for the vehicle, and a position and a speed for the mobile object are input to a first neural network that outputs an entry probability distribution of predicted entry times at which the mobile object will enter the target region. The target speed profile, the position and the speed for the vehicle, the position and the speed for the mobile object, and a predicted entry time at which the mobile object entered the target region are input to a second neural network that outputs an exit probability distribution of predicted exit times at which the mobile object will exit the target region. An optimized speed profile is determined based on the entry and exit probability distributions by executing a model predictive control algorithm in a rolling horizon. A vehicle is operated based on the optimized speed profile.

Abstract: A computer includes a processor and a memory storing instructions executable by the processor to receive a waypoint; generate a path for a vehicle from a starting point to the waypoint, the path characterized by a preset number of parameters; and instruct a propulsion and a steering system of the vehicle to actuate to navigate the vehicle along the path.

Abstract: A computer includes a processor and a memory, and the memory stores instructions executable by the processor to receive sensor data indicating an obstacle, formulate a first control barrier function for a vehicle based on the sensor data, formulate a second control barrier function for a trailer hitched to the vehicle based on the sensor data, determine a control input based on the first and second control barrier functions, and actuate a component of the vehicle according to the control input.

Abstract: A system for a vehicle can include a computer having a processor and a memory, the memory storing instructions executable by the processor, including instructions to determine a curvature command for a vehicle based on a feedforward control action based on a road curvature, a feedback control action based on at least one of a path offset or a heading offset of a vehicle with respect to a line defining a target lane; and to operate the vehicle with respect to the line defining the target lane based on the curvature command.

Abstract: A computer, including a processor and a memory, the memory including instructions to be executed by the processor to determine a first action based on inputting sensor data to a deep reinforcement learning neural network and transform the first action to one or more first commands. One or more second commands can be determined by inputting the one or more first commands to control barrier functions and transforming the one or more second commands to a second action. A reward function can be determined by comparing the second action to the first action. The one or more second commands can be output.

Abstract: A lateral virtual boundary for a host vehicle is identified based on a lateral distance between the host vehicle and a target vehicle, a longitudinal distance between the host vehicle and the target vehicle, and a speed of the target vehicle relative to the host vehicle. A forward virtual boundary for the host vehicle is identified based on the longitudinal distance between the host vehicle and the target vehicle. A lateral constraint value of the lateral virtual boundary and a forward constraint value of the forward virtual boundary are determined. A longitudinal acceleration and a steering angle are determined based on the lateral and forward virtual boundaries and the lateral and forward constraint values. One or both of a steering component or a brake are actuated based on the longitudinal acceleration and the steering angle.

Abstract: A control system and method for controlling a multiple gear ratio automatic transmission in a powertrain for an automatic transmission having pressure activated fiction torque elements to effect gear ratio upshifts. The friction torque elements are synchronously engaged and released during a torque phase of an upshift event as torque from a torque source is increased while allowing the off-going friction elements to slip, followed by an inertia phase during which torque from a torque source is modulated. A perceptible transmission output torque reduction during an upshift is avoided. Measured torque values are used during a torque phase of the upshift to correct an estimated oncoming friction element target torque so that transient torque disturbances at an oncoming clutch are avoided and torque transients at the output shaft are reduced.

Abstract: A vehicle hitch assistance system includes comprising an imaging system, a steering system, a dead reckoning device and a controller. The controller determines a hitch location within a vehicle exterior image received from the imaging system and controls the steering system to guide the vehicle in a reverse direction to align a hitch ball of the vehicle with the hitch location, including by tracking the location of the vehicle using information received from the dead reckoning device.

Abstract: A vehicle shock absorbing system includes a wheel, a vehicle body, a first absorber, a dynamic absorber, and a third absorber. The third absorber is attached to the vehicle body. The first absorber is between the third absorber and the wheel. The dynamic absorber is attached to the wheel and includes a dynamic absorber mass and a spring.

Abstract: A ride-sharing system may include a database configured to maintain a plurality of shuttle routes for ride-sharing and a server configured to receive a ride-sharing request indicating desire to join a ride-share and including a maximum walk distance to a pick-up or drop-off location of the desired route, and transmit, in response to the ride-sharing request, at least one selected route having a route walk distance to the pick-up or drop-off location not exceeding the maximum walk distance.

Abstract: A lane curvature, a vehicle position offset relative to a lane center line, and a vehicle heading are determined based on image data received from a vehicle camera sensor. An adjusted lane curvature and an adjusted vehicle heading are computed based on a vehicle speed, a vehicle yaw rate and the position offset.

Abstract: A computing system can determine a vehicle action based on inputting vehicle sensor data to a first neural network including a first safety agent that can determine a probability of unsafe vehicle operation. The first neural network can be adapted, at a plurality of times, by a periodically retrained deep reinforcement learning agent that includes a second deep neural network including a second safety agent. A vehicle can be operated based on the vehicle action.

Abstract: One or more target areas are identified proximate to a moving vehicle. The vehicle can be maneuvered to a target area selected according to a reinforcement learning reward function.

Abstract: A system includes a computer including a processor and a memory, the memory storing instructions executable by the processor to determine that a steering disturbance based on a difference between current vehicle state data and corresponding vehicle state model data exceeds a threshold, the vehicle state model including a yaw rate and a vehicle speed and, then, identify a wheel misalignment condition.

Abstract: Upon receiving a request to change lanes in a vehicle, a computer can determine that a first sightline toward a target lane is blocked, and can control the vehicle to move laterally in a current lane to obtain a second sightline toward the target lane.