Patents by Inventor Ramesh Sethu

Ramesh Sethu has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11455538
    Abstract: A method for reducing the number of neurons in a trained deep neural network (DNN) includes classifying layer types in a plurality of hidden layers; evaluating the accuracy of the DNN using a validation set of data; and generating a layer specific ranking of neurons, wherein the generating includes: analyzing, using the validation set of data for one or more of the plurality of hidden layers, the activation function for each neuron in the analyzed layers to determine an activation score for each neuron; and ranking, on a layer type basis, each neuron in the analyzed layers based on the neuron's activation score to generate a layer specific ranking of neurons. The method further includes removing a number of lower ranked neurons from the DNN that does not result in the DNN after the removal of selected lower ranked neurons to fall outside of an accuracy threshold limit.
    Type: Grant
    Filed: December 20, 2018
    Date of Patent: September 27, 2022
    Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Prakash Mohan Peranandam, Ramesh Sethu, Alena Rodionova
  • Publication number: 20220261627
    Abstract: Systems and method are provided for monitoring an operator of a vehicle. In one embodiment, a method includes: receiving, by a processor, data generated by the vehicle; determining, by the processor, causal time series event data based on the received data; computing, by the processor, a score for at least one of safety and quality based on a first machine learning model and the causal time series event data; computing, by the processor, at least one explanation for the score based on a second machine learning model; and generating, by the processor, display data to display at least one of the causal time series event data, the score, and the explanation to an end user.
    Type: Application
    Filed: February 12, 2021
    Publication date: August 18, 2022
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Ramesh Sethu, Prakash Mohan Peranandam, Rami I. Debouk, Arun Adiththan, Guangyu J. Zou
  • Publication number: 20200202214
    Abstract: A method for reducing the number of neurons in a trained deep neural network (DNN) includes classifying layer types in a plurality of hidden layers; evaluating the accuracy of the DNN using a validation set of data; and generating a layer specific ranking of neurons, wherein the generating includes: analyzing, using the validation set of data for one or more of the plurality of hidden layers, the activation function for each neuron in the analyzed layers to determine an activation score for each neuron; and ranking, on a layer type basis, each neuron in the analyzed layers based on the neuron's activation score to generate a layer specific ranking of neurons. The method further includes removing a number of lower ranked neurons from the DNN that does not result in the DNN after the removal of selected lower ranked neurons to fall outside of an accuracy threshold limit.
    Type: Application
    Filed: December 20, 2018
    Publication date: June 25, 2020
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Prakash Mohan Peranandam, Ramesh Sethu, Alena Rodionova
  • Patent number: 10534368
    Abstract: A crowdsourced virtual sensor generator is provided which could be generated by a vehicle or provided to the vehicle as, for example, a service. The crowdsourced virtual sensor generator may include, but is not limited to, a communication system configured to receive contributing vehicle sensor data from one or more contributing vehicles, a location of the one or more contributing vehicles and a target vehicle, and a processor, the processor configured to filter the received contributing vehicle sensor data based upon the location of the one or more contributing vehicles and the location of the target vehicle, aggregate the filtered contributing vehicle sensor data into at least one of a data-specific dataset and an application-specific data set, and generate a virtual sensor for the target vehicle, the virtual sensor processing the filtered and aggregated contributing vehicle sensor data to generate output data relative to the location of the target vehicle.
    Type: Grant
    Filed: June 26, 2017
    Date of Patent: January 14, 2020
    Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Ramesh Sethu, Arun Adiththan, Shuqing Zeng, Upali P. Mudalige
  • Publication number: 20180373266
    Abstract: A crowdsourced virtual sensor generator is provided which could be generated by a vehicle or provided to the vehicle as, for example, a service. The crowdsourced virtual sensor generator may include, but is not limited to, a communication system configured to receive contributing vehicle sensor data from one or more contributing vehicles, a location of the one or more contributing vehicles and a target vehicle, and a processor, the processor configured to filter the received contributing vehicle sensor data based upon the location of the one or more contributing vehicles and the location of the target vehicle, aggregate the filtered contributing vehicle sensor data into at least one of a data-specific dataset and an application-specific data set, and generate a virtual sensor for the target vehicle, the virtual sensor processing the filtered and aggregated contributing vehicle sensor data to generate output data relative to the location of the target vehicle.
    Type: Application
    Filed: June 26, 2017
    Publication date: December 27, 2018
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: RAMESH SETHU, ARUN ADITHTHAN, SHUQING ZENG, UPALI P. MUDALIGE
  • Publication number: 20180081681
    Abstract: A system, for use in reverse-engineering initial input initial code to a high-level equivalent model. The system includes a hardware-based processing unit and a non-transitory computer-readable storage component including a function-extraction module that, when executed by the hardware-based processing unit (i) generates, based on the input initial code and an input variable list, a list of output and state transition functions per task; and (ii) generates, based on an input task table, a scheduler-automaton structure. The storage component also includes a function-modeling module that, when executed, generates, using the scheduler automaton and the list of output and state transition functions per task, the high-level equivalent model of the input initial code. Various aspects of the present technology includes the non-transitory computer-readable storage devices configured to perform the operations described, and processes including the operations performed by these systems, storage devices, and algorithms.
    Type: Application
    Filed: September 16, 2016
    Publication date: March 22, 2018
    Inventor: Ramesh Sethu
  • Publication number: 20180060219
    Abstract: A system for use in analyzing software code using a symbolic-execution technique. The system includes a hardware-based processing unit, and a non-transitory computer-readable storage component including a (i) concurrency-analysis module, (ii) a lightweight-analysis module, and (iii) a heavyweight-analysis module. The concurrency-analysis module, when executed by the hardware-based processing unit receives initial code and generates a potential interference matrix using the initial code. The lightweight-analysis module, when executed by the hardware-based processing unit, generates a final interface matrix using the potential interference matrix. The heavyweight-analysis module, when executed by the hardware-based processing unit, generates one or more test cases using the potential interference matrix.
    Type: Application
    Filed: September 1, 2016
    Publication date: March 1, 2018
    Inventor: Ramesh Sethu
  • Publication number: 20160179868
    Abstract: A method of generating ontology models from requirement documents and software and performing consistency checks among requirement documents and software code utilizing ontology models. Terms in the plurality of requirement documents obtained from a database are identified. A processor assigns a part-of-speech tag to each term. The part-of-speech tag indicates a grammatical use of each term in the requirement documents. The processor classifies each term based on the part-of-speech tags. The classification identifies whether the each term is a part, symptom, action, event, or failure mode to constitute an ontology. The processor constructs an ontology-based consistency engine as a function of the ontologies. A consistency check is performed by applying the ontology-based consistency engine between ontologies extracted from two context documents. Inconsistent terms are identified between the context documents. At least one of the context documents having inconsistent terms is corrected.
    Type: Application
    Filed: December 18, 2014
    Publication date: June 23, 2016
    Inventors: Dnyanesh RAJPATHAK, Ramesh SETHU, Prakash M. PERANANDAM
  • Patent number: 9342489
    Abstract: A method of automatic identifying linking relationships of requirements in a plurality of requirement documents. Terms in the plurality of requirement documents are identified. A part-of-speech tag is assigned to each term. Each identified term is selected as a focal term. Co-occurring terms within a predetermined distance of the selected focal term are determined. A linking relationship probability is calculated for each co-occurring term associated with the selected focal term. The selected focal terms and associated co-occurring terms between the plurality of requirement documents are compared. A degree of linking relationship is identified between two requirements as a function of a comparison between selected focal terms and the associated co-occurring terms between the plurality of requirement documents. An analysis report identifying the degree of linking relationships between two respective requirements is output.
    Type: Grant
    Filed: April 3, 2014
    Date of Patent: May 17, 2016
    Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Ramesh Sethu, Prakash M. Peranandam, Dnyanesh Rajpathak, Soumen De
  • Patent number: 9342441
    Abstract: A method of establishing traceability for embedded software systems. A design code database is provided for an embedded software system. A test suite database including a plurality of test cases is structured for testing design code of the embedded software system. The structuring of the test cases provides a correspondence from a respective test case to a respective portion of the design code. A processor receives a design code modification to the embedded software. An associated test case is identified for testing the modified design code being based on traceability data. The associated test case is revised to accommodate the modified design code. The modified test cases are integrated into the test suite. A traceability database establishes a one-to-one correspondence between the modified design coder and the modified test case is updated.
    Type: Grant
    Filed: June 3, 2014
    Date of Patent: May 17, 2016
    Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Ramesh Sethu, Shige Wang
  • Publication number: 20150347279
    Abstract: A method of establishing traceability for embedded software systems. A design code database is provided for an embedded software system. A test suite database including a plurality of test cases is structured for testing design code of the embedded software system. The structuring of the test cases provides a correspondence from a respective test case to a respective portion of the design code. A processor receives a design code modification to the embedded software. An associated test case is identified for testing the modified design code being based on traceability data. The associated test case is revised to accommodate the modified design code. The modified test cases are integrated into the test suite. A traceability database establishes a one-to-one correspondence between the modified design coder and the modified test case is updated.
    Type: Application
    Filed: June 3, 2014
    Publication date: December 3, 2015
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: RAMESH SETHU, SHIGE WANG
  • Publication number: 20150286631
    Abstract: A method of automatic identifying linking relationships of requirements in a plurality of requirement documents. Terms in the plurality of requirement documents are identified. A part-of-speech tag is assigned to each term. Each identified term is selected as a focal term. Co-occurring terms within a predetermined distance of the selected focal term are determined. A linking relationship probability is calculated for each co-occurring term associated with the selected focal term. The selected focal terms and associated co-occurring terms between the plurality of requirement documents are compared. A degree of linking relationship is identified between two requirements as a function of a comparison between selected focal terms and the associated co-occurring terms between the plurality of requirement documents. An analysis report identifying the degree of linking relationships between two respective requirements is output.
    Type: Application
    Filed: April 3, 2014
    Publication date: October 8, 2015
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: RAMESH SETHU, PRAKASH M. PERANANDAM, DNYANESH RAJPATHAK, SOUMEN DE
  • Patent number: 9152385
    Abstract: Systems and methods for generating formal software requirements using an informal requirements document having informal requirements and annotations associated with the informal requirements. The systems and methods extract syntax from the annotations and generate artifacts as a function of the syntax.
    Type: Grant
    Filed: February 22, 2012
    Date of Patent: October 6, 2015
    Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Arun Chakrapani Rao, Manoj G. Dixit, Ramesh Sethu
  • Patent number: 8683442
    Abstract: A method allows for testing software under test (SUT) with respect to a partial design model (PDM) having a boundary which differs from a boundary of the SUT. The method includes recording input information including the SUT, the PDM, and coverage criteria defining a required number of the test cases. Variables in the SUT are identified that correspond to boundary signals for the PDM. Test cases are extracted meeting the coverage criteria. The method may include generating additional test cases at the PDM level and mapping the additional cases with corresponding constraint functions to the boundary of the SUT using a forward/backward propagation and/or heuristics guided technique. A system for testing the SUT includes a host machine and memory. The host machine executes process instructions from memory to identify variables in the SUT that correspond to boundary signals for the PDM, and extracts test cases meeting the coverage criteria.
    Type: Grant
    Filed: September 21, 2011
    Date of Patent: March 25, 2014
    Assignee: GM Global Technology Operations LLC
    Inventors: Prakash Mohan Peranandam, Ambar A. Gadkari, Ramesh Sethu
  • Patent number: 8589898
    Abstract: A computer-implemented method for evaluating a machine-executable software code specification includes using the computer to generate a system dependence graph corresponding to the software code specification. The system dependence graph includes elements including nodes and edges. The computer evaluates the system dependence graph including selecting a variable modified in the software code specification, providing a control operation node of the system dependence graph corresponding to a control statement in the software code specification with a preferred calibration state, traversing to selected elements of the system dependence graph wherein the selected elements are associated with the selected variable and the preferred calibration state of the control operation node, evaluating only the selected elements of the system dependence graph, and identifying ones of the selected elements whereat a state of the selected variable is modified.
    Type: Grant
    Filed: March 29, 2010
    Date of Patent: November 19, 2013
    Assignee: GM Global Technology Operations LLC
    Inventors: Srihari Sukumaran, Ramesh Sethu
  • Patent number: 8584108
    Abstract: A computer-implemented method for evaluating a machine-executable software code specification includes using the computer to generate a system dependence graph corresponding to the software code specification. The system dependence graph includes elements including nodes and edges, wherein the computer evaluates the system dependence graph. The evaluation of the system dependence graph includes selecting a variable modified in the software code specification, traversing to selected elements of the system dependence graph, the selected elements associated with the selected variable, evaluating only the selected elements of the system dependence graph, and identifying ones of the selected elements whereat a state of the selected variable is modified.
    Type: Grant
    Filed: March 29, 2010
    Date of Patent: November 12, 2013
    Assignee: GM Global Technology Operations LLC
    Inventors: Srihari Sukumaran, Ramesh Sethu
  • Publication number: 20130219354
    Abstract: Systems and methods for generating formal software requirements using an informal requirements document having informal requirements and annotations associated with the informal requirements. The systems and methods extract syntax from the annotations and generate artifacts as a function of the syntax.
    Type: Application
    Filed: February 22, 2012
    Publication date: August 22, 2013
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Arun Chakrapani Rao, Manoj G. Dixit, Ramesh Sethu
  • Patent number: 8484009
    Abstract: A method and tools for providing precise timing analysis scalable to industrial case studies with large numbers of tasks and messages are provided, including the capability to model and analyze task and message response times; ECU usage; bus usage; end-to-end latency of task/message chains; and timing synchronization problems in task/message graphs. System tasks and messages are modeled in a formalism known as calendar automaton. Models are written in a modeling language such as Promela and instrumented with code specific to the analysis specification. Models and instrumentation are automatically generated from the system description and analysis specification. The system model is subjected to exhaustive state space exploration by a compatible model checker, such as SPIN. During exploration, the instrumented code produces results for different timing analyses.
    Type: Grant
    Filed: May 14, 2010
    Date of Patent: July 9, 2013
    Assignee: GM Global Technology Operations LLC
    Inventors: Swarup K. Mohalik, Rajeev Ariyattu Chandrasekharan, Manoj G. Dixit, Ramesh Sethu, Devesh B Chokshi
  • Patent number: 8453119
    Abstract: A system and method for automatic formal verification of an executable model includes an assertion monitor configured to verify a system against an assertion in a specification. The assertion monitor includes a parser configured to generate a propositional formula representing the assertion in the specification using Boolean propositions, a filter configured to generate a run of the system using truth assignments for the propositional symbols, and a trace verifier configured to verify the assertion using the run of the system using truth assignments for the propositional symbols and the propositional formula.
    Type: Grant
    Filed: October 14, 2009
    Date of Patent: May 28, 2013
    Assignee: GM Global Technology Operations LLC
    Inventors: Swarup K. Mohalik, Prasanna Vignesh V. Ganesan, Ramesh Sethu
  • Publication number: 20130074040
    Abstract: A method allows for testing software under test (SUT) with respect to a partial design model (PDM) having a boundary which differs from a boundary of the SUT. The method includes recording input information including the SUT, the PDM, and coverage criteria defining a required number of the test cases. Variables in the SUT are identified that correspond to boundary signals for the PDM. Test cases are extracted meeting the coverage criteria. The method may include generating additional test cases at the PDM level and mapping the additional cases with corresponding constraint functions to the boundary of the SUT using a forward/backward propagation and/or heuristics guided technique. A system for testing the SUT includes a host machine and memory. The host machine executes process instructions from memory to identify variables in the SUT that correspond to boundary signals for the PDM, and extracts test cases meeting the coverage criteria.
    Type: Application
    Filed: September 21, 2011
    Publication date: March 21, 2013
    Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC
    Inventors: Prakash Mohan Peranandam, Ambar A. Gadkari, Ramesh Sethu