公开(公告)号：US20240362924A1

公开(公告)日：2024-10-31

申请号：US18645882

申请日：2024-04-25

申请人： Tata Consultancy Services Limited

发明人： Gireesh NANDIRAJU ,  Ayush AGRAWAL ,  Ahana DATTA ,  Snehasis BANERJEE ,  Mohan SRIDHARAN ,  Madhava KRISHNA ,  Brojeshwar BHOWMICK

IPC分类号： G06V20/58 ,  G05D1/246 ,  G05D1/633 ,  G05D1/644 ,  G05D101/00 ,  G05D101/15 ,  G06T7/246 ,  G06V10/82

CPC分类号： G06V20/58 ,  G05D1/246 ,  G05D1/633 ,  G05D1/644 ,  G06T7/246 ,  G06V10/82 ,  G05D2101/15 ,  G05D2101/22 ,  G06T2207/20084 ,  G06T2207/30261

摘要： This disclosure relates generally to method and system for multi-object tracking and navigation without pre-sequencing. Multi-object navigation is an embodied Al task where object navigation only searches for an instance of at least one target object where a robot localizes an instance to locate target objects associated with an environment. The method of the present disclosure employs a deep reinforcement learning (DRL) based framework for sequence agnostic multi-object navigation. The robot receives from an actor critic network a deterministic local policy to compute a low-level navigational action to navigate along a shortest path calculated from a current location of the robot to the long-term goal to reach the target object. Here, a deep reinforcement learning network is trained to assign the robot with a computed reward function when the navigational action is performed by the robot to reach an instance of the plurality of target objects.

公开(公告)号：US20240362470A1

公开(公告)日：2024-10-31

申请号：US18479893

申请日：2023-10-03

申请人： MACRONIX INTERNATIONAL CO., LTD.

发明人： Yu-Chen LU ,  Sheng-Feng YU ,  Wei-Cheng LIN ,  Chi-Chih CHANG ,  Pei-Shuo WANG ,  Kuan-Cheng LIN ,  Kai-Chiang WU

IPC分类号： G06N3/0495 ,  G06N3/08 ,  G06V20/58

CPC分类号： G06N3/0495 ,  G06N3/08 ,  G06V20/58

摘要： The application provides a panoramic perception method, system and a non-transitory computer readable medium. The panoramic perception method comprises: performing a first pretraining on a plurality of weights of a training model using the source database; performing a second pretraining with data augmentation on the plurality of weights of the training model using the source database; performing a combined training on the plurality of weights of the training model using both the source database and the target database; performing a quantization-aware training on the plurality of weights of the training model using the source database and the target database; performing a post training quantization on the plurality of weights of the training model using the target database; and performing panoramic perception by the training model.

公开(公告)号：US12131620B2

公开(公告)日：2024-10-29

申请号：US17455732

申请日：2021-11-19

申请人： Robert Bosch GmbH

发明人： Aimee Williams ,  Christopher Prediger ,  Danielle Heinz ,  Steffen Kutter ,  Jia Hou ,  Sylvia Briese ,  Camille Vuillaume

IPC分类号： H04W4/90 ,  A61B5/024 ,  A61B5/0245 ,  G06V20/58 ,  G06V40/10 ,  G06V40/20 ,  G08B21/02 ,  G08B25/01 ,  G08B25/10 ,  H04W4/02 ,  H04W4/12 ,  H04W4/40 ,  H04W4/80

CPC分类号： G08B25/016 ,  A61B5/02438 ,  A61B5/02455 ,  G06V20/58 ,  G06V40/10 ,  G06V40/20 ,  G08B21/02 ,  G08B25/10 ,  H04W4/023 ,  H04W4/027 ,  H04W4/12 ,  H04W4/90 ,  H04W4/40 ,  H04W4/80

摘要： A method of operating a personal safety system including a vehicle includes automatically detecting that a smartphone is within a communication range using a wireless communication system of the vehicle, and starting an approach time period with a controller of the vehicle or the smartphone when the smartphone is detected. The method further includes sending checkup data from the vehicle to the smartphone using the wireless communication system when the smartphone fails to arrive at the vehicle prior to the approach time period elapsing, and triggering an emergency response by the vehicle and/or the smartphone when the vehicle fails to receive a response to the checkup data from the smartphone within a response time period.

公开(公告)号：US12131562B2

公开(公告)日：2024-10-29

申请号：US18079498

申请日：2022-12-12

申请人： Motional AD LLC

发明人： Sergi Adipraja Widjaja ,  Dhananjai Sharma ,  Venice Erin B. Liong

IPC分类号： G01C21/32 ,  G01S7/48 ,  G01S17/42 ,  G01S17/86 ,  G01S17/89 ,  G01S17/931 ,  G06T7/10 ,  G06V10/80 ,  G06V10/82 ,  G06V20/58 ,  G06V20/70 ,  H04N23/80

CPC分类号： G06V20/70 ,  G01S7/4802 ,  G01S17/42 ,  G01S17/86 ,  G01S17/89 ,  G01S17/931 ,  G06T7/10 ,  G06V10/806 ,  G06V10/82 ,  G06V20/58 ,  H04N23/80 ,  G06T2207/20084

摘要： Provided are methods for enhanced semantic labeling in mapping with a semantic labeling system, which can include receiving, from a LiDAR sensor of a vehicle, LiDAR point cloud information including at least one raw point feature for a point, receiving, from a camera of the vehicle, image data associated with an image captured using the camera, generating at least one rich point feature for the point based on the image data, predicting, using a LiDAR segmentation neural network and based on the at least one raw point feature and the at least one rich point feature, a point-level semantic label for the point, and providing the point-level semantic label to a mapping engine to generate a map based on the point-level semantic label Systems and computer program products are also provided.

公开(公告)号：US12131554B2

公开(公告)日：2024-10-29

申请号：US17728376

申请日：2022-04-25

申请人： Ningbo Geely Automobile Research & Development Co., Ltd.

发明人： Erik Lindberg Nilsson ,  Jonathan Johansson

IPC分类号： G06V20/58 ,  G06T7/70 ,  G06V10/764 ,  G06V10/774 ,  G06V10/776 ,  G06V10/80 ,  G06V20/40 ,  H04N5/77

CPC分类号： G06V20/58 ,  G06T7/70 ,  G06V10/764 ,  G06V10/7747 ,  G06V10/776 ,  G06V10/811 ,  G06V20/41 ,  H04N5/77 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/30261

摘要： An object detection method includes obtaining a first video stream having video images with object detection data of an area, identifying an object in the area by identifying one or more first object detection data of the object in the first video stream corresponding to validated object detection data, determining a position of the object in the area, obtaining a second video stream having video images with object detection data of the area, identifying second object detection data of the second video stream at the position of the object and determining whether the second object detection data is un-validated object detection data, and updating the validated object detection data of the object detection system if it is determined that the second object detection data of the object is un-validated object detection data.

公开(公告)号：US12131553B2

公开(公告)日：2024-10-29

申请号：US17623160

申请日：2020-05-20

申请人： Volkswagen Aktiengesellschaft

发明人： Stephan Max ,  Martin Wegner ,  Sabina Alazzawi ,  Richard Niestroj

IPC分类号： G06V20/58 ,  G06T7/10

CPC分类号： G06V20/58 ,  G06T7/10

摘要： The disclosure relates to a method, to a computer program with instructions, and to a device for processing data recorded by a motor vehicle. The disclosure additionally relates to a motor vehicle and to a back end in which a method according to the disclosure or a device according to the disclosure is used. In a first step, an item of data recorded by a motor vehicle is received. A temporal obfuscation is then applied to the received item of data. In the process, a temporally segmented obfuscation interval is used for the temporal obfuscation. The obfuscated item of data is finally forwarded for further processing. The temporal obfuscation can be carried out within the motor vehicle or in a back end connected to the motor vehicle.

公开(公告)号：US12131552B2

公开(公告)日：2024-10-29

申请号：US17593355

申请日：2020-03-17

申请人： SONY GROUP CORPORATION

发明人： Ayaka Hirata ,  Kei Nakagawa

IPC分类号： G06V20/58 ,  G01W1/10 ,  G06V10/28

CPC分类号： G06V20/58 ,  G01W1/10 ,  G06V10/28

摘要： An imaging system of the present disclosure includes: an event detection device that detects an event; and a controller that controls the event detection device. The controller then controls the detection sensitivity of event detection being performed by the event detection device, on the basis of external information. Further, an object recognition system of the present disclosure includes: an event detection device that detects an event; a controller that controls the detection sensitivity of event detection being performed by the event detection device, on the basis of external information; and a recognition processing unit that performs object recognition in an event, on the basis of an event signal output from the event detection device.

公开(公告)号：US12131551B2

公开(公告)日：2024-10-29

申请号：US17567023

申请日：2021-12-31

申请人： Rivian IP Holdings, LLC

发明人： Gary Fay ,  Krishna Prasad Varadarajan Srinivasan

IPC分类号： G06V20/58 ,  B60W40/105 ,  G06T7/20 ,  G06T7/70 ,  G06V10/25 ,  G06V10/74

CPC分类号： G06V20/58 ,  B60W40/105 ,  G06T7/20 ,  G06T7/70 ,  G06V10/25 ,  G06V10/761 ,  B60W2554/80 ,  G06T2207/20081 ,  G06T2207/30252 ,  G06T2210/12

摘要： Systems and methods are provided to receive, at a processor associated with a vehicle and via one or more image sensors associated with the vehicle, image data associated with an environment surrounding the vehicle and corresponding to a first image captured at a first time, and additional image data associated with an environment surrounding the vehicle and corresponding to a second image captured by at a second time. The provided systems and methods may determine, based on the received additional image data and a machine learning model, that a tracked object identified in the first image is not detected in the second image, and may determine, based on vehicle data and tracking data of the tracked object, that the tracked object should be present in the second image and perform a remedial action on the additional image data to identify the tracked object in the second image.

公开(公告)号：US20240355129A1

公开(公告)日：2024-10-24

申请号：US18380009

申请日：2023-10-13

申请人： HL Klemove Corp.

发明人： Eunsan JO ,  Jun-Hyung CHA

IPC分类号： G06V20/56 ,  B60W40/072 ,  B60W50/14 ,  G06V20/58

CPC分类号： G06V20/588 ,  B60W40/072 ,  B60W50/14 ,  G06V20/582 ,  B60W2552/30 ,  B60W2552/53 ,  B60W2555/60 ,  B60W2556/10 ,  B60W2556/20

摘要： A method for correcting lane recognition information using road traffic sign recognition information, and an apparatus for performing the method according to various exemplary embodiments of the present disclosure verify a validity of lane recognition information based on lane recognition information and independent road traffic sign recognition information, and correct the lane recognition information when the lane recognition information is information which is not valid, thereby enhancing lane recognition performance used for various advanced driver assistance systems (ADAS) functions.

公开(公告)号：US20240355128A1

公开(公告)日：2024-10-24

申请号：US18753361

申请日：2024-06-25

申请人： TOYOTA RESEARCH INSTITUTE, INC.

发明人： PEIYAN GONG ,  Michael J. Delp ,  Christian Potthast

IPC分类号： G06V20/58 ,  B60W40/04 ,  G06F18/22 ,  G06T7/10 ,  G06T7/90 ,  G06T17/05

CPC分类号： G06V20/584 ,  B60W40/04 ,  G06F18/22 ,  G06T7/10 ,  G06T7/90 ,  G06T17/05 ,  B60W2555/60 ,  G06T2207/10024 ,  G06T2207/30252

摘要： A system and method are provided for making a probabilistic determination of the state of a lane transition which may be controlled by a traffic signal, such as a traffic cue. A determination of the state of transition of a traffic signal may be used by a vehicle to determine a course of action to take. In making the determination, the states of multiple traffic signals may be combined into one collection of elements that has values associated with the likelihood of the traffic signals being in a given state. Optionally, a determination may be made of whether a traffic signal is occluded.