Patents by Inventor Sai Bhargav Yalamanchi
Sai Bhargav Yalamanchi 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: 11851087Abstract: Systems, methods, tangible non-transitory computer-readable media, and devices associated with trajectory prediction are provided. For example, trajectory data and goal path data can be accessed. The trajectory data can be associated with an object's predicted trajectory. The predicted trajectory can include waypoints associated with waypoint position uncertainty distributions that can be based on an expectation maximization technique. The goal path data can be associated with a goal path and include locations the object is predicted to travel. Solution waypoints for the object can be determined based on application of optimization techniques to the waypoints and waypoint position uncertainty distributions. The optimization techniques can include operations to maximize the probability of each of the solution waypoints. Stitched trajectory data can be generated based on the solution waypoints. The stitched trajectory data can be associated with portions of the solution waypoints and the goal path.Type: GrantFiled: July 26, 2022Date of Patent: December 26, 2023Assignee: UATC, LLCInventors: Nemanja Djuric, Sai Bhargav Yalamanchi, Galen Clark Haynes, Tzu-Kuo Huang
-
Patent number: 11762391Abstract: Systems and methods for training machine-learned models are provided. A method can include receiving a rasterized image associated with a training object and generating a predicted trajectory of the training object by inputting the rasterized image into a first machine-learned model. The method can include converting the predicted trajectory into a rasterized trajectory that spatially corresponds to the rasterized image. The method can include utilizing a second machine-learned model to determine an accuracy of the predicted trajectory based on the rasterized trajectory. The method can include determining an overall loss for the first machine-learned model based on the accuracy of the predictive trajectory as determined by the second machine-learned model. The method can include training the first machine-learned model by minimizing the overall loss for the first machine-learned model.Type: GrantFiled: September 9, 2022Date of Patent: September 19, 2023Assignee: UATC, LLCInventors: Henggang Cui, Junheng Wang, Sai Bhargav Yalamanchi, Mohana Prasad Sathya Moorthy, Fang-Chieh Chou, Nemanja Djuric
-
Publication number: 20230122535Abstract: A method includes receiving a digital surface model of an area for unmanned aerial vehicle (UAV) navigation. The digital surface model represents an environmental surface in the area. The method includes determining, for each grid cell of a plurality of grid cells in the area, a confidence value of an altitude of the environmental surface at the grid cell and determining a terrain clearance value based at least on the confidence value of the altitude of the environmental surface at the grid cell. The method includes determining a route for a UAV through the area such that the altitude of the UAV is above the altitude of the environmental surface at each grid cell of a sequence of grid cells of the route by at least the terrain clearance value determined for the grid cell. The method includes causing the UAV to navigate through the area using the determined route.Type: ApplicationFiled: October 19, 2022Publication date: April 20, 2023Inventors: Dinuka Abeywardena, Konstantin Bozhkov, Kyle Kakligian, Stephen Lacy, Scott Barron, Brandon Jones, Aditya Undurti, Kyle David Julian, Sai Bhargav Yalamanchi
-
Publication number: 20230021034Abstract: Systems and methods for training machine-learned models are provided. A method can include receiving a rasterized image associated with a training object and generating a predicted trajectory of the training object by inputting the rasterized image into a first machine-learned model. The method can include converting the predicted trajectory into a rasterized trajectory that spatially corresponds to the rasterized image. The method can include utilizing a second machine-learned model to determine an accuracy of the predicted trajectory based on the rasterized trajectory. The method can include determining an overall loss for the first machine-learned model based on the accuracy of the predictive trajectory as determined by the second machine-learned model. The method can include training the first machine-learned model by minimizing the overall loss for the first machine-learned model.Type: ApplicationFiled: September 9, 2022Publication date: January 19, 2023Inventors: Henggang Cui, Junheng Wang, Sai Bhargav Yalamanchi, Mohana Prasad Sathya Moorthy, Fang-Chieh Chou, Nemanja Djuric
-
Publication number: 20220388537Abstract: Systems, methods, tangible non-transitory computer-readable media, and devices associated with trajectory prediction are provided. For example, trajectory data and goal path data can be accessed. The trajectory data can be associated with an object's predicted trajectory. The predicted trajectory can include waypoints associated with waypoint position uncertainty distributions that can be based on an expectation maximization technique. The goal path data can be associated with a goal path and include locations the object is predicted to travel. Solution waypoints for the object can be determined based on application of optimization techniques to the waypoints and waypoint position uncertainty distributions. The optimization techniques can include operations to maximize the probability of each of the solution waypoints. Stitched trajectory data can be generated based on the solution waypoints. The stitched trajectory data can be associated with portions of the solution waypoints and the goal path.Type: ApplicationFiled: July 26, 2022Publication date: December 8, 2022Inventors: Nemanja Djuric, Sai Bhargav Yalamanchi, Galen Clark Haynes, Tzu-Kuo Huang
-
Patent number: 11442459Abstract: Systems and methods for training machine-learned models are provided. A method can include receiving a rasterized image associated with a training object and generating a predicted trajectory of the training object by inputting the rasterized image into a first machine-learned model. The method can include converting the predicted trajectory into a rasterized trajectory that spatially corresponds to the rasterized image. The method can include utilizing a second machine-learned model to determine an accuracy of the predicted trajectory based on the rasterized trajectory. The method can include determining an overall loss for the first machine-learned model based on the accuracy of the predictive trajectory as determined by the second machine-learned model. The method can include training the first machine-learned model by minimizing the overall loss for the first machine-learned model.Type: GrantFiled: February 6, 2020Date of Patent: September 13, 2022Assignee: UATC, LLCInventors: Henggang Cui, Junheng Wang, Sai Bhargav Yalamanchi, Mohana Prasad Sathya Moorthy, Fang-Chieh Chou, Nemanja Djuric
-
Patent number: 11420648Abstract: Systems, methods, tangible non-transitory computer-readable media, and devices associated with trajectory prediction are provided. For example, trajectory data and goal path data can be accessed. The trajectory data can be associated with an object's predicted trajectory. The predicted trajectory can include waypoints associated with waypoint position uncertainty distributions that can be based on an expectation maximization technique. The goal path data can be associated with a goal path and include locations the object is predicted to travel. Solution waypoints for the object can be determined based on application of optimization techniques to the waypoints and waypoint position uncertainty distributions. The optimization techniques can include operations to maximize the probability of each of the solution waypoints. Stitched trajectory data can be generated based on the solution waypoints. The stitched trajectory data can be associated with portions of the solution waypoints and the goal path.Type: GrantFiled: February 29, 2020Date of Patent: August 23, 2022Assignee: UATC, LLCInventors: Nemanja Djuric, Sai Bhargav Yalamanchi, Galen Clark Haynes, Tzu-Kuo Huang
-
Publication number: 20210269059Abstract: Systems, methods, tangible non-transitory computer-readable media, and devices associated with trajectory prediction are provided. For example, trajectory data and goal path data can be accessed. The trajectory data can be associated with an object's predicted trajectory. The predicted trajectory can include waypoints associated with waypoint position uncertainty distributions that can be based on an expectation maximization technique. The goal path data can be associated with a goal path and include locations the object is predicted to travel. Solution waypoints for the object can be determined based on application of optimization techniques to the waypoints and waypoint position uncertainty distributions. The optimization techniques can include operations to maximize the probability of each of the solution waypoints. Stitched trajectory data can be generated based on the solution waypoints. The stitched trajectory data can be associated with portions of the solution waypoints and the goal path.Type: ApplicationFiled: February 29, 2020Publication date: September 2, 2021Inventors: Nemanja Djuric, Sai Bhargav Yalamanchi, Galen Clark Haynes, Tzu-Kuo Huang
-
Publication number: 20210181754Abstract: Systems and methods for training machine-learned models are provided. A method can include receiving a rasterized image associated with a training object and generating a predicted trajectory of the training object by inputting the rasterized image into a first machine-learned model. The method can include converting the predicted trajectory into a rasterized trajectory that spatially corresponds to the rasterized image. The method can include utilizing a second machine-learned model to determine an accuracy of the predicted trajectory based on the rasterized trajectory. The method can include determining an overall loss for the first machine-learned model based on the accuracy of the predictive trajectory as determined by the second machine-learned model. The method can include training the first machine-learned model by minimizing the overall loss for the first machine-learned model.Type: ApplicationFiled: February 6, 2020Publication date: June 17, 2021Inventors: Henggang Cui, Junheng Wang, Sai Bhargav Yalamanchi, Mohana Prasad Sathya Moorthy, Fang-Chieh Chou, Nemanja Djuric