A misfire detection device for an internal combustion engine is provided. A mapping takes time series data of instantaneous speed parameters as inputs. Each instantaneous speed parameter corresponds to one of a plurality of successive second intervals in a first interval. The instantaneous speed parameters correspond to the rotational speed of the crankshaft. The first interval is a rotational angular interval of the crankshaft in which compression top dead center occurs. The second interval is smaller than an interval between compression top dead center positions. The mapping outputs a probability that a misfire has occurred in at least one cylinder that reaches compression top dead center in the first interval. The mapping data defining the mapping has been learned by machine learning.

A controller controls a vehicle including an engine with a fuel vapor processing device. The fuel vapor processing device executes purge control that sends fuel vapor of a fuel tank, via a canister, to an intake passage on condition that air-fuel ratio learning is complete. The controller includes processing circuitry. The processing circuitry automatically stops the engine when an automatic stopping condition is satisfied, automatically starts the engine when an automatic starting condition is satisfied, determines that a prohibition condition for prohibiting automatic stopping is satisfied when the air-fuel ratio learning is incomplete, and inhibits automatic stopping of the engine even if the automatic stopping condition is satisfied when determining that the prohibition condition is satisfied.

An engine optimization controller may determine values of a set of input parameters of an engine and process, using an engine model, the values to determine a plurality of sets of potential output parameters. The engine optimization controller may determine, based on the plurality of sets of potential output parameters, an engine optimization scheme, which may indicate a first number of cylinders, of one or more cylinders of the engine, to be active and to receive gas; a second number of cylinders, of the one or more cylinders, to be inactive and to receive gas, and/or a third number of cylinders, of the one or more cylinders, to be inactive and to not receive gas. The engine optimization controller may provide the engine configuration scheme to another controller to allow control of the one or more cylinders and one or more fuel injectors according to the engine configuration scheme.

An engine controller includes an ignition timing control unit and a rich imbalance determining unit. The rich imbalance determining unit designates one of multiple cylinders as a subject cylinder for determination and executes lean active control that commands a smaller amount of fuel injection for the subject cylinder than for the other cylinders. The rich imbalance determining unit determines whether an air-fuel ratio of the subject cylinder deviates to be richer based on a rotational fluctuation amount of a crankshaft during the execution of the lean active control. The ignition timing control unit executes an advancement limiting process that limits advancement of the ignition timing by the knock control during the execution of the lean active control.

A method of adaptively predicting, during operation of a pump, a mass of fuel pumped by the pump during a pumping event to a fuel accumulator (“Q.sub.pump”) to control operation of the pump is provided, comprising: generating an adaptive model of operation of the pump, including estimating a start of pumping (“SOP”) position of a plunger of the pump, estimating Q.sub.pump, determining a converged value of the estimated SOP position, and determining a converged value of the estimated Q.sub.pump; using the adaptive model to predict Q.sub.pump by inputting to the model the converged value of the estimated SOP position, a measured pressure of fuel in the fuel accumulator and a measured temperature of fuel in the fuel accumulator; and controlling operation of the pump in response to the predicted Q.sub.pump.

A state estimation device for an internal combustion engine includes: a storage device that stores mapping data, the mapping data being data defining a mapping that takes as an input an internal combustion engine state variable and that generates as an output an estimated value for estimating the state of the internal combustion engine; and an execution device that executes an acquisition process of acquiring the internal combustion engine state variable and an estimation process of calculating the estimated value based on the output of the mapping. The mapping data is data learned by machine learning. When the estimated value is out of an acceptable range, the execution device executes a guard process of adjusting the estimated value to a value close to or within the acceptable range. When executing the guard process, the execution device calculates the value after the guard process as the estimated value.

A novel control device for an internal combustion engine capable of highly accurately estimating an EGR amount (rate) during the transient state is provided. A first EGR rate is determined using, as an input, a detection signal of an EGR sensor provided on the downstream side of a throttle valve which adjusts the flow rate of a mixed gas of air and EGR gas flowing through an intake pipe, a second EGR rate is estimated by calculating a predetermined equation using, as an input, at least a detection signal of an air flow sensor and an EGR valve opening degree sensor, a third EGR rate is determined by carrying out delay processing on the second EGR rate corresponding to a response delay of the EGR sensor, and the second EGR rate is subjected to learning correction by reflecting a difference between the third EGR rate and the first EGR rate.

A misfire detection device for an internal combustion engine is provided. A mapping takes time series data of instantaneous speed parameters as inputs. Each instantaneous speed parameter corresponds to one of a plurality of successive second intervals in a first interval. The instantaneous speed parameters correspond to the rotational speed of the crankshaft. The first interval is a rotational angular interval of the crankshaft in which compression top dead center occurs. The second interval is smaller than an interval between compression top dead center positions. The mapping outputs a probability that a misfire has occurred in at least one cylinder that reaches compression top dead center in the first interval. The mapping data defining the mapping has been learned by machine learning.

A control system of an internal combustion engine comprises an air-fuel ratio sensor 40, 41 detecting an air-fuel ratio of exhaust gas, a current detecting device 61 detecting an output current of the air-fuel ratio sensor, a voltage applying device 60 applying voltage to the air-fuel ratio sensor, and a voltage control part 81 configured to control voltage applied to the air-fuel ratio sensor through the voltage applying device. The voltage control part is configured to set the applied voltage to a reference voltage determined so that the output current becomes zero when an air-fuel ratio of inflowing exhaust gas flowing into the air-fuel ratio sensor is a stoichiometric air-fuel ratio, and correct the reference voltage so that the output current detected by the current detecting device becomes zero when it is judged that the air-fuel ratio of the inflowing exhaust gas is the stoichiometric air-fuel ratio.

A vehicle control data generation method includes causing processing circuitry to execute an obtaining process that obtains a state of a vehicle and a specifying variable, an operating process that operates an electronic device, a reward calculating process that provides a greater reward when a characteristic of the vehicle meets a standard than when the characteristic does not meet the standard, and an updating process that updates relationship defining data. The update map outputs the updated relationship defining data. The reward calculating process includes a changing process that changes the reward, provided when the characteristic of the vehicle is a predetermined characteristic, such that the reward in a case where torque generated by an internal combustion engine is used to generate the propelling force of the vehicle differs from the reward in a case where the torque is not used to generate the propelling force.