-
公开(公告)号：EP3772707A1
公开(公告)日：2021-02-10
申请号：EP19190105.7
申请日：2019-08-05
Applicant: Robert Bosch GmbH , CARNEGIE MELLON UNIVERSITY
Inventor： Kolter, Jeremy Zieg , Manek, Gaurav , Vinogradska, Julia
Abstract: A system and computer-implemented method are provided for training a dynamics model to learn the dynamics of a physical system. In particular, the dynamics model may be learned to be able to infer a future state of the physical system and/or its environment based on a current state of the physical system and/or its environment. The learned dynamics model is inherently globally stable. Namely, instead of learning a dynamics model and attempting to separately verify its stability, the learnable dynamics model comprises a learnable Lyapunov function which is jointly learned together with the nominal dynamics of the physical system. Accordingly, the learned dynamics model is highly suitable for real-life applications in which a physical system may assume a state which was unseen during training as the learned dynamics model is inherently globally stable.
-
公开(公告)号：EP3885848A1
公开(公告)日：2021-09-29
申请号：EP20165352.4
申请日：2020-03-24
Applicant: Robert Bosch GmbH , Carnegie Mellon University
Inventor： Kolter, Jeremy Zieg , Donti, Priya L. , Fazlyab, Mahyar , Vinogradska, Julia , Roderick, Melrose
IPC: G05B13/02
Abstract: Some embodiments are directed to a controller for generating a control signal for a computer-controlled machine. A neural network may be applied to a current sensor signal, the neural network being configured to map the sensor signal to a raw control signal. A projection function may be applied to the raw control signal to obtain a stable control signal to control the computer-controllable machine.
-
Patent Agency Ranking
公开(公告)号：EP3796108A1
公开(公告)日：2021-03-24
申请号：EP19198558.9
申请日：2019-09-20
Applicant: Robert Bosch GmbH
Inventor： Klenske, Edgar , Froehlich, Lukas , Vinogradska, Julia , Zeilinger, Melanie
Abstract: Eine Vorrichtung und ein Verfahren zum Ermitteln eines robusten Optimums eines physikalischen oder chemischen Prozesses gemäß einem Bayes-Optimierungsverfahren werden offenbart, wobei jeder Messpunkt von bekannten Messpunkten (202) einen Eingangsparameterwert eines physikalischen oder chemischen Prozesses und einen dem Eingangsparameterwert zugeordneten gemessenen Ausgangsparameterwert aufweist, wobei ein erstes statistisches Modell den Zusammenhang zwischen den Eingangsparameterwerten und den Ausgangsparameterwerten des physikalischen oder chemischen Prozesses beschreibt, und wobei bei dem Verfahren ein zweites statistisches Modell (208) für das erste statistische Modell (204) ermittelt wird, wobei das zweite statistische Modell (208) eine Robustheit für die Ausgangsparameterwerte des physikalischen oder chemischen Prozesses bezüglich einer Veränderung der Eingangsparameterwerte (206) beschreibt, und wobei ein neuer Messpunkt (210) derart ausgewählt wird, dass eine Differenz aus einer Entropie des durch die bekannten Messpunkte (202) beschriebenen ersten statistischen Modells (204) für den physikalischen oder chemischen Prozess an dem neuen Messpunkt (210) und einer erwarteten Entropie des durch die bekannten Messpunkte (202) beschriebenen ersten statistischen Modells (204) an dem neuen Messpunkt (210) und einem angenommenen Maximalwert des Ausgangsparameters des zweiten statistischen Modells (208) an dem robusten Optimum des physikalischen oder chemischen Prozesses im Wesentlichen maximal ist oder in einem vorgegebenen Umgebungsbereich des Maximums liegt.
-
4.发明公开
公开(公告)号：EP4502735A1
公开(公告)日：2025-02-05
申请号：EP23200195.8
申请日：2023-09-27
Applicant: Robert Bosch GmbH
Inventor： Vinogradska, Julia , Peters, Jan , Berkenkamp, Felix , Bottero, Alessandro Giacomo , Luis Goncalves, Carlos Enrique
IPC: G05B13/02
Abstract: The invention relates to a computer-implemented control method (700) of constrained controlling of a computer-controlled system. The system is controlled according to a control input, which is safe if a constraint quantity resulting from the controlling of the computer-controlled system exceeds a constraint threshold. A current control input is determined based on previous control inputs and corresponding previous noisy measurements. The computer-controlled system is controlled according to the current control input, thereby obtaining a current noisy measurement of the resulting constraint quantity. The current control input is determined based on a mutual information between a first random variable representing the constraint quantity resulting from the current control input and a second random variable indicating whether a further control input is safe and based on the Max-value entropy search (MES) acquisition function.
-
公开(公告)号：EP4307055A1
公开(公告)日：2024-01-17
申请号：EP22184158.8
申请日：2022-07-11
Applicant: Robert Bosch GmbH
Inventor： Vinogradska, Julia , Peters, Jan , Berkenkamp, Felix , Bottero, Alessandro Giacomo , Luis Goncalves, Carlos Enrique
Abstract: The invention relates to a computer-implemented control method (700) of constrained controlling of a computer-controlled system. The system is controlled according to a control input, which is safe if a constraint quantity resulting from the controlling of the computer-controlled system exceeds a constraint threshold. A current control input is determined based on previous control inputs and corresponding previous noisy measurements. The computer-controlled system is controlled according to the current control input, thereby obtaining a current noisy measurement of the resulting constraint quantity. The current control input is determined based on a mutual information between a first random variable representing the constraint quantity resulting from the current control input and a second random variable indicating whether a further control input is safe.
-
公开(公告)号：EP4502870A1
公开(公告)日：2025-02-05
申请号：EP23189842.0
申请日：2023-08-04
Applicant: Robert Bosch GmbH
Inventor： Vinogradska, Julia , Peters, Jan , Berkenkamp, Felix , Bottero, Alessandro Giacomo , Luis Goncalves, Carlos Enrique
Abstract: According to various embodiments, a method for training a control policy for controlling a technical system is described, comprising controlling the technical system and observing state transitions of the technical system in response to controlling the technical system and observing rewards received from the state transitions, adapting a transition dynamic distribution model modelling a distribution of the dynamics of the technical system to the observed state transitions, sampling a transition dynamic from the transition dynamic distribution model, updating a value function distribution model modelling a distribution of a value function of states or state-action pairs of the technical system by estimating the value of a state using the sampled transition dynamic and updating the control strategy using the updated value function distribution model.
-
公开(公告)号：EP4386632A1
公开(公告)日：2024-06-19
申请号：EP22213403.3
申请日：2022-12-14
Applicant: Robert Bosch GmbH
Inventor： Vinogradska, Julia , Peters, Jan , Berkenkamp, Felix , Bottero, Alessandro Giacomo , Luis Goncalves, Carlos Enrique
Abstract: According to various embodiments, a method for training a control policy is described, comprising estimating the variance of a value function which associates a state with a value of the state or a pair of state and action with a value of the pair by solving a Bellman uncertainty equation, wherein, for each of multiple states, the reward function of the Bellman uncertainty equation is set to the difference of the total uncertainty about the mean of the value of the subsequent state following the state and the average aleatoric uncertainty of the value of the subsequent state and biasing the control policy in training towards regions for which the estimation gives a higher variance of the value function than for other regions.
-
-
-
-
-
-
Search Results
7
records
(took 0.013 seconds)
