Robert Boschgmbh

Google Inc.

Joerg Mueller

Jason Scott Hardy

Mithun Jacob

Oliver F Schwindt

Ammar Husain

Thomas Glaser

Theresa Veronika Kienle

Reiner Kunz

Jan Kleindieck

Andrei Vatavu

Simon Ehmann

Seyed Abbas Sadat

Jingru Luo

A method and device are disclosed for improved object detection in an area surrounding a robot. In the method, first and second sensing data are obtained, which can be assigned to a first or second sensing means of the robot, respectively, and which contain at least one portion of the area surrounding the robot. An objection detection of an object in the area surrounding the robot is carried out using a fusion of at least the first and the second sensing data. An item of redundancy information is generated, which is assigned to the object detection and at least indicates whether the detected object has been detected using only the first or only the second sensing data or whether the detected object or at least one or more sections of same has been detected redundantly using both the first and the second sensing data.

Method and device for operating a robot with improved object detection

Implementations set forth herein relate to generating training data, such that each instance of training data includes a corresponding instance of vision data and drivability label(s) for the instance of vision data. A drivability label can be determined using first vision data from a first vision component that is connected to the robot. The drivability label(s) can be generated by processing the first vision data using geometric and/or heuristic methods. Second vision data can be generated using a second vision component of the robot, such as a camera that is connected to the robot. The drivability labels can be correlated to the second vision data and thereafter used to train one or more machine learning models. The trained models can be shared with a robot(s) in furtherance of enabling the robot(s) to determine drivability of areas captured in vision data, which is being collected in real-time using one or more vision components.

Generating and/or using training instances that include previously captured robot vision data and drivability labels

An action planning system () and method for autonomous vehicles are provided. The system () comprises one or more processors () and one or more non-transitory computer-readable storage medium () having stored thereon a computer program used by the one or more processors (), wherein the computer program causes the one or more processors () to estimate future environment of an autonomous vehicle (), generate a possible trajectory for the autonomous vehicle (), predict motion and reactions of each dynamic obstacle in the future environment of the autonomous vehicle () based on current local traffic context, and generate a prediction iteratively over timesteps.

Action planning system and method for autonomous vehicles

A system for a vehicle, which drives in an at least in-part-automated manner is configured to determine a control signal for a control system. The system includes a sensor, a planning module, and a monitoring module. The sensor is configured to detect an object in a surrounding area of the vehicle and store a corresponding object representation. The planning module is configured to determine, based to the stored object representation, a first trajectory and a first probability of collision of the first trajectory for the vehicle. The monitoring module is configured to perform one of following actions when the first probability of collision exceeds a predefined probability of collision: determine, using the planning module and based on the stored object representation, a further trajectory having a further probability of collision and a maximum deceleration of the further trajectory; or assess the stored object representation of the object using the sensor.

Determination of a control signal for an in-part-autonomous vehicle

A method of operating a vehicle includes determining a first plurality of known lane intervals at a first time with a controller of a guidance system of the vehicle, the known lane intervals available to receive the vehicle, determining a plurality of visible lane intervals at a second time after the first time with the controller based on perception data generated by the guidance system, and determining a plurality of occupied lane intervals at the second time with the controller based on the perception data generated by the guidance system. The method further includes determining a second plurality of known lane intervals at the second time as a union of the first plurality of known lane intervals, the plurality of visible lane intervals, and the plurality of occupied lane intervals.

Known lane intervals for automated driving

A processing system includes at least one computer processor, which is configured to implement a method that includes receiving crowd-sourced driving path data, map data, and real-time sensor data associated with a roadway. The method includes generating, by the processing system, optimized driving path data for the roadway based on the crowd-sourced path data, the map data, and the sensor data. The method includes providing control signals to control an autonomous or semi-autonomous vehicle based on the optimized driving path data.

System and method of using crowd-sourced driving path data in an autonomous or semi-autonomous driving system

A hydraulic two-way valve includes a pump connection, a tank connection, and first and second working connections. The working connections are connectable to a hydraulic cylinder of an agricultural vehicle lifting mechanism. The pump connection is connectable to a hydraulic pump. In a first position, the pump connection, tank connection, and working connections are shut off. In a second position, the pump connection is connected to the first working connection and the tank connection to the second working connection. In a third position, the pump connection is connected to the second working connection and the tank connection to the first working connection. In a fourth position, the working connections are connected to the tank connection and the pump connection is shut off. In a fifth position, the tank connection is connected to the second working connection and the pump connection and first working connection are shut off.

Growing community of inventors

Mountain View, CA, United States of America

Joerg Mueller