The present invention is a method for determining an optimal speed of a first vehicle preceded by a second vehicle. Position, speed and acceleration of the second vehicle are measured in order to determine a trajectory thereof, and a dynamic model of the first vehicle is constructed. The optimal speed is then determined by minimizing the energy consumption of the vehicle by use of the dynamic model by minimization being constrained by the trajectory of the second vehicle.

A controller and a method for controlling motion of a vehicle is provided. The method comprises acquiring motion information including a current state of the vehicle and a desired state of the vehicle, determining a combination of a steering angle of the wheels and motor forces for moving the vehicle from the current state into the desired state by using a first model of the motion of the vehicle and a second model of the motion of the chassis of the vehicle, determining a cost function of the motion of the vehicle, optimizing the cost function of the motion of the vehicle to compute a command signal for controlling the steering wheel and the plurality of electric motors, and controlling the steering angle of the wheels and the motor forces based on the command signal.

The subject disclosure relates to techniques for minimizing damage for collisions including an autonomous vehicle. A process of the disclosed technology can include predicting that a crash involving the autonomous vehicle is imminent, altering at least one operational parameter of the autonomous vehicle after predicting the crash is imminent, performing a first simulation on the autonomous vehicle, wherein the first simulation is a simulation of the autonomous vehicle taking a first action to minimize damage from the crash, and generating a first damage estimate for the first simulation.

A method for calculating a management setpoint for managing fuel and electric power consumption of a hybrid motor vehicle. The method includes: a) acquiring, via a navigation system, a journey to be made, b) dividing the journey into successive sections, c) acquiring, for each section, attributes characterizing the section, d) deducing a relation linking the estimated fuel consumption of the hybrid motor vehicle over the section to the estimated electric power consumption of the hybrid motor vehicle, e) measuring the actual fuel consumption and the actual electric power consumption of the motor vehicle, f) correcting the relation, taking into account the actual fuel and electric power consumptions, and g) determining an optimal consumption point in each of the corrected relations in order to minimize the fuel consumption of the motor vehicle over the journey as a whole.

The operation of a hybrid powertrain system is optimized with respect to a desired state-of-charge trajectory, taking account of the estimated anticipated vehicle drive power. The hybrid powertrain system has an internal combustion engine and an electrically operated torque machine. The internal combustion engine and the torque machine are controlled by a control device and are connected to an output element via a hybrid transmission. Before the start of the prediction period Δt, an experience-based state-of-charge trajectory for the anticipated route, covering at least the prediction period Δt, is retrieved from an external database. The desired state-of-charge trajectory is established based on the experience-based state-of-charge trajectory by modifying it with at least one optimization constraint. The experience-based state-of-charge trajectory can be established based on operating data from hybrid powertrain systems of multiple vehicles and/or from operating data from multiple comparable journeys with the same vehicle.

Techniques for utilizing a Monte Carlo model to perform pre-training of a ground vehicle controller. A sampled distribution of actions and corresponding states can be utilized to train a reinforcement learning controller policy, learn an action-value function, or select a set of control parameters with a predetermined loss.

According to an embodiment, a system generates a number of sample trajectories from a trajectory sample space for a driving scenario. The system determines a reward based on a reward model for each of the sample trajectories, where the reward model is generated using a rank based conditional inverse reinforcement learning algorithm. The system ranks the sample trajectories based on the determined rewards. The system determines a highest ranked trajectory based on the ranking. The system selects the highest ranked trajectory to control the ADV autonomously according to the highest ranked trajectory.

A ground vehicle control system including a plurality of sensors and one or more predictive controllers. The sensors can be configured to detect environment information and vehicle operating information. The one or more predictive controllers can be configured to self-train for an energy consumption solution based on one or more control parameters including the environment information and the vehicle operating information.

Systems and methods for generating motion plans for autonomous vehicles are provided. An autonomous vehicle can include a machine-learned motion planning system including one or more machine-learned models configured to generate target trajectories for the autonomous vehicle. The model(s) include a behavioral planning stage configured to receive situational data based at least in part on the one or more outputs of the set of sensors and to generate behavioral planning data based at least in part on the situational data and a unified cost function. The model(s) includes a trajectory planning stage configured to receive the behavioral planning data from the behavioral planning stage and to generate target trajectory data for the autonomous vehicle based at least in part on the behavioral planning data and the unified cost function.

A method for determining a trajectory of a motor vehicle includes identifying a plurality of objects present in the surroundings of the motor vehicle. For each object, the method includes: a) determining a speed of impact between the object of interest and the motor vehicle, b) determining a risk of injury in the event of a collision with the motor vehicle at the determined impact speed, c) determining the probability of a collision resulting in an injury between the object of interest and the motor vehicle, depending on the determined risk of injury. The method subsequently includes determining a plurality of possible trajectories for the motor vehicle, and determining the trajectory to be followed by the motor vehicle by optimising a cost function which depends on the determined collision probabilities and which minimises the risk of collision resulting in an injury between each object and the motor vehicle.