A method for determining a state of a transmission for a vehicle includes providing an input for a first generative model depending on a route information, a vehicle speed, a probabilistic variable, and an output of a second physical model, and determining an output of the first model characterizing the state of the transmission in response to the input for the first model. The first model includes a first layer trained to map input to an intermediate state and a second layer trained to map the intermediate state to the state of the transmission depending on the output of the second model. The method includes providing an input for the second physical model depending on at least one vehicle state and/or the route information, and determining an output of the second model in response to the input for the second model. The output of the second model characterizes limit(s) for the intermediate state.

A method for the dynamically expanding play correction according to a method for hysteresis compensation for an actuator and for a shift fork which is movable by this actuator via an electric motor having a rotor and a stator and which guides a gearshift sleeve, by means of a cellular automaton, wherein a torque ripple of the actuator and a mechanical displacement of the gearshift sleeve are compensated independently of one another or in combination by means of a learning algorithm.

A dynamic belt tension inference method includes steps of: (A) performing a training process and (B) performing an inference process. Step (A) includes steps of: using Isolation Forest algorithm to train a tension inference model; using Isolation Forest algorithm under the same hyper-parameter set to perform multiple trainings to generate multiple tension inference models; respectively computing multiple model performances of the multiple tension inference models according to the anomaly score and a pre-recorded data label; computing an averaged model performance; determine whether multiple averaged model performances have been acquired; selecting one of the multiple hyper-parameter sets that corresponds to an optimal averaged model performance as a final hyper-parameter set for training to output an final model. Step (B) includes a step of: inferring processed data by the final model to generate the anomaly score and the dynamic tension corresponding to the anomaly score.

A transmission which includes input and output shafts and first and second step-down devices for producing various step-down ratios. The first and second step-down devices are associated with first and second controllable clutches, respectively, in order to connect, in a torque-transmitting manner, the respective step-down device between the input shaft and the output shaft. A first gear of the transmission is obtained by actuating the first clutch and a second gear by actuating the second clutch. A method of controlling a gearshift, from the first gear to the second gear, includes the steps of determining operating condition parameters of the transmission; and determining optimized degrees of actuation of the clutches by a search process related to a transmission model that determines the degrees of actuation on the basis of operating condition parameters in relation to a predetermined optimization criterion. The transmission model is based on slip.

A dynamic belt tension inference method includes steps of: (A) performing a training process and (B) performing an inference process. Step (A) includes steps of: using Isolation Forest algorithm to train a tension inference model; using Isolation Forest algorithm under the same hyper-parameter set to perform multiple trainings to generate multiple tension inference models; respectively computing multiple model performances of the multiple tension inference models according to the anomaly score and a pre-recorded data label; computing an averaged model performance; determine whether multiple averaged model performances have been acquired; selecting one of the multiple hyper-parameter sets that corresponds to an optimal averaged model performance as a final hyper-parameter set for training to output an final model. Step (B) includes a step of: inferring processed data by the final model to generate the anomaly score and the dynamic tension corresponding to the anomaly score.

A system includes a torque converter having a turbine, a transmission having friction clutches and an input member connected to the turbine, and a controller. The controller is programmed to control a change-of-mind shift maneuver of the transmission. By executing a method, the controller detects the change-of-mind shift maneuver, predicts an acceleration profile of the turbine for a next-requested shift of the detected shift maneuver as a function of a calibrated desired shift time and an output speed of the transmission, calculates a shift control value for a next-requested shift of the shift maneuver using the predicted acceleration profile, and executes the next-requested shift via the clutches using the calculated shift control value. The shift maneuver may be a skip-at-sync quick shift-to-quick shift or power downshift-to-power downshift. The shift value may be a clutch pressure for an offgoing holding clutch or a torque management level from an engine.

A method of operating a transmission includes measuring an output torque, estimating a gearbox input torque using a model, and estimating gearbox component torques based on a detailed gearbox model. The model used to estimate the input torque varies depending on whether a torque converter is locked, open, or slipping. In some operating conditions, multiple estimates are available for gearbox input torque, impeller torque, or shift element torque in which case the models are adapted. When an estimated component torque is outside an expected range, a warning flag is raised and diagnostic data is saved. When an estimated torque approaches or exceeds a torque limit, the input torque command may be reduced to prevent component damage. A warning flag may also be raised and diagnostic data saved in response to a model parameter being adapted to a value outside of a predetermined range.

A control unit includes an excessive output suppression control unit which suppresses an excessive output by reducing a current command value from a host control apparatus. A control signal output unit obtains a motor control signal by adding a reduction amount of the current command value to the current command value. The excessive output suppression control unit includes: an oil pressure estimating unit which estimates oil pressure based on a current and a rotating speed of the electric motor; and a current command value correction amount calculating unit which compares outputs the reduction amount of the current command value if the estimated oil pressure is higher than target oil pressure. One of the target oil pressure and the estimated oil pressure compared by the current command value correction amount calculating unit is corrected based on oil temperature information.

A system includes a torque converter having a turbine, a transmission having friction clutches and an input member connected to the turbine, and a controller. The controller is programmed to control a change-of-mind shift maneuver of the transmission. By executing a method, the controller detects the change-of-mind shift maneuver, predicts an acceleration profile of the turbine for a next-requested shift of the detected shift maneuver as a function of a calibrated desired shift time and an output speed of the transmission, calculates a shift control value for a next-requested shift of the shift maneuver using the predicted acceleration profile, and executes the next-requested shift via the clutches using the calculated shift control value. The shift maneuver may be a skip-at-sync quick shift-to-quick shift or power downshift-to-power downshift. The shift value may be a clutch pressure for an offgoing holding clutch or a torque management level from an engine.