Patents by Inventor Ayush Shah
Ayush Shah 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: 11977854Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: GrantFiled: April 17, 2023Date of Patent: May 7, 2024Assignee: UNLIKELY ARTIFICIAL INTELLIGENCE LIMITEDInventors: William Tunstall-Pedoe, Robert Heywood, Seth Warren, Paul Benn, Duncan Reynolds, Ayush Shah, Luci Krnic, Ziyi Zhu
-
Publication number: 20240095468Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: ApplicationFiled: November 21, 2023Publication date: March 21, 2024Inventors: William TUNSTALL-PEDOE, Robert HEYWOOD, Seth WARREN, Paul BENN, Duncan REYNOLDS, Ayush SHAH, Luci KRNIC, Ziyi ZHU
-
Publication number: 20240066706Abstract: Various embodiments are directed to methods, apparatuses, systems, computing devices, computing entities, and the like for robotically depalletizing objects. In various embodiments, a controller-implemented method for robotic depalletization may comprise detecting a first object of a plurality of palletized objects arranged in a plurality of pallet layers, the first object defining at least a portion of a top pallet layer; identifying an initial pallet height defined by the first object; identifying a second pallet height defined by a second object of the plurality of palletized objects; detecting a pallet height difference based at least in part on the second pallet height; and determining that the top pallet layer has been depalletized based at least in part on a comparison of the pallet height difference to a threshold pallet layer height.Type: ApplicationFiled: August 30, 2022Publication date: February 29, 2024Inventors: Abhijit MAKHAL, Ayush JHALANI, Darshan SHAH
-
Publication number: 20230316006Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: ApplicationFiled: April 17, 2023Publication date: October 5, 2023Inventors: William TUNSTALL-PEDOE, Robert HEYWOOD, Seth WARREN, Paul DARIAS, Duncan REYNOLDS, Ayush SHAH, Luci KRNIC, Ziyi ZHU
-
Patent number: 11772517Abstract: A transportation system is disclosed. The transportation system includes a vehicle, a plurality of seating assemblies positioned within a passenger compartment of the vehicle and defining an arrangement, a plurality of actuators that effect movement of various components of the plurality of seating assemblies, a plurality of sensors, and a controller. Specific examples of pre-set or pre-programmed arrangements are disclosed, as well as the ability to customize the arrangement. Additionally, exemplary methods are disclosed that illustrate transitions between a variety of the pre-set or pre-programmed arrangements.Type: GrantFiled: November 9, 2020Date of Patent: October 3, 2023Assignee: Ford Global Technologies, LLCInventors: Marcos Silva Kondrad, Patrick Maloney, Kevin Mozurkewich, Johnathan Andrew Line, Ayush Shah
-
Publication number: 20230274094Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: ApplicationFiled: April 17, 2023Publication date: August 31, 2023Inventors: William TUNSTALL-PEDOE, Robert HEYWOOD, Seth WARREN, Paul DARIAS, Duncan REYNOLDS, Ayush SHAH, Luci KRNIC, Ziyi ZHU
-
Publication number: 20230274086Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: ApplicationFiled: April 17, 2023Publication date: August 31, 2023Inventors: William TUNSTALL-PEDOE, Robert HEYWOOD, Seth WARREN, Paul DARIAS, Duncan REYNOLDS, Ayush SHAH, Luci KRNIC, Ziyi ZHU
-
Publication number: 20230274089Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: ApplicationFiled: April 17, 2023Publication date: August 31, 2023Inventors: William TUNSTALL-PEDOE, Robert HEYWOOD, Seth WARREN, Paul DARIAS, Duncan REYNOLDS, Ayush SHAH, Luci KRNIC, Ziyi ZHU
-
Publication number: 20230259705Abstract: Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.Type: ApplicationFiled: April 17, 2023Publication date: August 17, 2023Inventors: William TUNSTALL-PEDOE, Robert HEYWOOD, Seth WARREN, Paul DARIAS, Duncan REYNOLDS, Ayush SHAH, Luci KRNIC, Ziyi ZHU
-
Publication number: 20230181868Abstract: The disclosure is generally directed to providing a relaxation regimen to an occupant of a vehicle. In one example method, a processor evaluates a physiological condition of the occupant. The physiological condition can be a heart rate, hear-rate variability, a breathing rate, and/or a blood oxygen level. The processor also identifies an opportune time to offer the relaxation regimen. The relaxation regimen may be offered based on evaluating the physiological condition of the occupant and on identifying the opportune time. Furthermore, the relaxation regimen may be recommended based on the customer preference or usage patterns and the opportune time. The opportune time can be when the vehicle is moving in a cruising mode of operation, is in a stopped or parked state, or is moving in stop-and-go traffic, or location. In another example method, a processor may offer the relaxation regime based on a movement status of the vehicle.Type: ApplicationFiled: December 9, 2021Publication date: June 15, 2023Applicant: Ford Global Technologies, LLCInventors: Kwaku Prakah-Asante, Ayush Shah
-
Patent number: 11675828Abstract: A method, a computer program product, and a computer system determine and arrange images to include in a visual representation. The method includes receiving a textual statement and identifying a plurality of terms in the textual statement that are to be visualized in the visual representation. The method includes generating a plurality of sequences of images where each image in a given one of the sequences is associated with one of the terms. Each image is associated with at least one tag. The method includes determining a global coherence and a local coherence for each of the sequences based on the tags of the images. The method includes selecting one of the sequences based on the global coherence and the local coherence. The method includes generating the visual representation where the images of the selected sequence are included.Type: GrantFiled: November 5, 2019Date of Patent: June 13, 2023Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Tamer E. Abuelsaad, Ayush Shah
-
Patent number: 11626016Abstract: A method is disclosed for vehicular operation on a road comprising a road network configured to communicate with one or more vehicles on the road, the method comprising sending a signal from a first vehicle to the road network, the signal comprising an intended path of the first vehicle on the road; receiving the signal at the road network; and adjusting at least one of a speed and a path of a second vehicle with respect to the road in response to the signal.Type: GrantFiled: March 15, 2019Date of Patent: April 11, 2023Assignee: Ford Global Technologies, LLCInventors: Oliver Lei, Perry MacNeille, Ayush Shah
-
Patent number: 11507758Abstract: Vehicle-based sign language communication systems and methods are provided herein. An example device can be configured to determine a sign language protocol used by the first user, determine a target language used by a second user, obtain a translation library based on the sign language protocol and the target language, receive spoken word input from a second user through a microphone, convert the spoken word input into sign language output using the translation library, and provide the sign language output using a sign language output device.Type: GrantFiled: October 30, 2019Date of Patent: November 22, 2022Assignee: Ford Global Technologies, LLCInventors: Omar Makke, Oleg Gusikhin, Ayush Shah
-
Publication number: 20220144135Abstract: A transportation system is disclosed. The transportation system includes a vehicle, a plurality of seating assemblies positioned within a passenger compartment of the vehicle and defining an arrangement, a plurality of actuators that effect movement of various components of the plurality of seating assemblies, a plurality of sensors, and a controller. Specific examples of pre-set or pre-programmed arrangements are disclosed, as well as the ability to customize the arrangement. Additionally, exemplary methods are disclosed that illustrate transitions between a variety of the pre-set or pre-programmed arrangements.Type: ApplicationFiled: November 9, 2020Publication date: May 12, 2022Applicant: Ford Global Technologies, LLCInventors: Marcos Silva Kondrad, Patrick Maloney, Kevin Mozurkewich, Johnathan Andrew Line, Ayush Shah
-
Publication number: 20210325520Abstract: Systems and methods for calibrating a LiDAR device are disclosed. According to one embodiment, the system comprises a LiDAR device, a continuous curved target at a fixed distance from the LiDAR device, and a calibration controller operable to perform a reflectance over range calibration of the LiDAR device. The LiDAR device scans portions of the continuous curved target at different ranges during the calibration.Type: ApplicationFiled: April 17, 2020Publication date: October 21, 2021Applicant: VELODYNE LIDAR, INC.Inventors: Zhongping Cai, Geovany Ramirez, Ayush Shah, Shing Leung Luk, Marcus Smith, David S. Hall
-
Patent number: 11136730Abstract: A method includes positioning one or more roadway sensors and a plurality of a networked array of light emitting diode (LED) raised pavement markers on a road. Sensors proximate the road are associated with a network of road marking controllers that cooperate to control the LEDs as vehicles drive on the road. The LEDs operate as a networked array of roadway lane marking lights. The road marking controller(s) determine a distance between a vehicle on the road and the roadway sensor, determine a dynamic condition associated with the vehicle that changes with respect to time, and lights the plurality of LEDs based at least in part on the distance between the vehicle and the roadway sensor and the dynamic condition associated with the vehicle. Dynamic conditions for operating the LED lighting scheme can include vehicle velocity, a date, a time, a weather condition proximate the vehicle, and other factors.Type: GrantFiled: September 24, 2019Date of Patent: October 5, 2021Assignee: Ford Global Technologies, LLCInventors: Ayush Shah, Oliver Lei, Perry Robinson MacNeille
-
Patent number: 11077860Abstract: Systems and methods are disclosed for an embedded in-vehicle platform that provides immersive user experiences. An example method includes obtaining, by a vehicle controller of a vehicle, a preference profile for a user; loading the preference profile by the vehicle controller; determining a shopping experience of the user occurring on a vehicle shopping platform within the vehicle; determining an ambience profile of a merchant identified in the shopping experience; and controlling, by the vehicle controller, an in-vehicle ambience within the vehicle based on the ambience profile and the preference profile.Type: GrantFiled: March 7, 2019Date of Patent: August 3, 2021Assignee: Ford Global Technologies, LLCInventors: Ayush Shah, Jeffrey Yeung, Harald C. Martinez, Oleg Gusikhin
-
Patent number: 11055333Abstract: A processor extracts a sentence from a portion of text. The sentence includes one or more words. A concreteness score is determined for each word in the sentence. A set of concrete words is determined based upon a comparison of the concreteness score for each word and a predetermined threshold. A grammatical dependency relationship is determined between the one or more words of the sentence. One or more subsets of search terms is determined based upon the grammatical dependency relationship. Each member of the one or more subsets of search terms is a member of the set of concrete words. One or more images is retrieved from a repository based on the one or more subsets of search terms.Type: GrantFiled: January 8, 2019Date of Patent: July 6, 2021Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Tamer E. Abuelsaad, Ayush Shah
-
Publication number: 20210133232Abstract: A method, a computer program product, and a computer system determine and arrange images to include in a visual representation. The method includes receiving a textual statement and identifying a plurality of terms in the textual statement that are to be visualized in the visual representation. The method includes generating a plurality of sequences of images where each image in a given one of the sequences is associated with one of the terms. Each image is associated with at least one tag. The method includes determining a global coherence and a local coherence for each of the sequences based on the tags of the images. The method includes selecting one of the sequences based on the global coherence and the local coherence. The method includes generating the visual representation where the images of the selected sequence are included.Type: ApplicationFiled: November 5, 2019Publication date: May 6, 2021Inventors: Tamer E. Abuelsaad, Ayush Shah
-
Publication number: 20210133290Abstract: Vehicle-based sign language communication systems and methods are provided herein. An example device can be configured to determine a sign language protocol used by the first user, determine a target language used by a second user, obtain a translation library based on the sign language protocol and the target language, receive spoken word input from a second user through a microphone, convert the spoken word input into sign language output using the translation library, and provide the sign language output using a sign language output device.Type: ApplicationFiled: October 30, 2019Publication date: May 6, 2021Applicant: Ford Global Technologies, LLCInventors: Omar Makke, Oleg Gusikhin, Ayush Shah