A dock assembly is provided. The dock assembly is configured for docking with a robot. An alignment platform of said dock assembly is configured to receive a sweeper module from the robot when the robot is docked and said sweeper module disengages from the robot. The alignment platform has a plurality of cones positioned on a top side of the alignment platform. The plurality of cones are configured to engage a plurality of holes positioned on an underside of the sweeper module when the sweeper module becomes disengaged from the robot. The plurality of cones enable self-alignment of the alignment platform to the sweeper module as the plurality of cones engage the plurality of holes. The alignment platform has a plurality of support pads positioned on a bottom side of the alignment platform. The support pads are configured to rest on a plurality of bearings that permit lateral movement of the alignment platform when the plurality of cones engage the plurality of holes and the alignment platform self-aligns to the sweeper module.

An autonomous off-road vehicle, upon encountering an obstruction while navigating a route, can apply a first machine-learned model to identify the obstruction. In the event that the first machine-learned model cannot identify the obstruction, the autonomous off-road vehicle can apply a second machine-learned model configured to determine whether or not the obstruction can be ignored, for instance based on dimensions of the obstruction. If the obstruction can be ignored, the autonomous off-road vehicle can continue navigating the route. If the obstruction cannot be ignored, the autonomous off-road vehicle can modify the route, can stop, can flag the obstruction to a remote human operator, can modify an interface of a human operator to display a notification or a video feed from the vehicle, and the like.

Techniques to use a trained model to determine a yaw of an object are described. For example, a system may implement various techniques to generate multiple representations for an object in an environment. Each representation vary based on the technique and data used. An estimation component may estimate a representation from the multiple representations. The model may be implemented to output a yaw for the object using the multiple representations, the estimated representation, and/or additional information. The output yaw may be used to track an object, generate a trajectory, or otherwise control a vehicle.

Techniques to use a trained model to determine a yaw of an object are described. For example, a system may implement various techniques to generate multiple representations for an object in an environment. Each representation vary based on the technique and data used. An estimation component may estimate a representation from the multiple representations. The model may be implemented to output a yaw for the object using the multiple representations, the estimated representation, and/or additional information. The output yaw may be used to track an object, generate a trajectory, or otherwise control a vehicle.

Disclosed herein are systems and methods for autonomous vehicle operation. A computing system can include a communication device configured to receive a plurality of event signals from at least a first autonomous vehicle that is traversing a path, and a processor in electrical communication with the communication device and configured to determine whether the event signals are indicative of an obstacle in a portion of the path. The communication device can be configured to receive, from at least a second autonomous vehicle, at least one characteristic of the obstacle captured by at least one sensor of the second autonomous vehicle, and transmit, to at least a third autonomous vehicle, at least one task to clear the obstacle from the portion of the path. The processor can be configured to determine, based on the characteristic of the obstacle, the at least one task to be transmitted by the communication device.

Disclosed are techniques for estimating a 3D bounding box (3DBB) from a 2D bounding box (2DBB). Conventional techniques to estimate 3DBB from 2DBB rely upon classifying target vehicles within the 2DBB. When the target vehicle is misclassified, the projected bounding box from the estimated 3DBB is inaccurate. To address such issues, it is proposed to estimate the 3DBB without relying upon classifying the target vehicle.

A robot including a medium storing instructions that when executed by a processor of the robot effectuates operations including: capturing images of a workspace as the robot moves within the workspace; identifying at least one characteristic of an object captured in the images of the workspace; determining an object type of the object based on an object dictionary of different types of objects, wherein the different object types comprise at least a cord, clothing garments, a shoe, earphones, and pet bodily waste; and instructing the robot to execute at least one action based on the object type of the object, wherein the at least one action comprises avoiding the object or cleaning around the object.

A machine-learning (ML) architecture for determining three or more outputs, such as a two and/or three-dimensional region of interest, semantic segmentation, direction logits, depth data, and/or instance segmentation associated with an object in an image. The ML architecture may output these outputs at a rate of 30 or more frames per second on consumer grade hardware.

A method includes maneuvering a robot in (i) a following mode in which the robot is controlled to travel along a path segment adjacent an obstacle, while recording data indicative of the path segment, and (ii) in a coverage mode in which the robot is controlled to traverse an area. The method includes generating data indicative of a layout of the area, updating data indicative of a calculated robot pose based at least on odometry, and calculating a pose confidence level. The method includes, in response to the confidence level being below a confidence limit, maneuvering the robot to a suspected location of the path segment, based on the calculated robot pose and the data indicative of the layout and, in response to detecting the path segment within a distance from the suspected location, updating the data indicative of the calculated pose and/or the layout.

A control system, a control method, and a program capable of lowering difficulty of a user's work on an autonomous mobile robot having a placement part are provided. A control system for controlling an autonomous mobile robot including a placement part on which a load is placed includes a user recognition unit configured to recognize a user of the placement part, a feature information acquisition unit configured to acquire feature information of the recognized user, and an operation control unit configured to control a height of the placement part based on the feature information.