Aspects of the disclosure relate to controlling an autonomous vehicle to respond to a detected collision. An autonomous vehicle control system may receive sensor data associated with an autonomous vehicle in which the autonomous vehicle control system is installed. The autonomous vehicle control system may analyze the sensor data in real-time as the sensor data is received and may detect an occurrence of a collision involving the autonomous vehicle. In response to detecting the occurrence of the collision, the autonomous vehicle control system may generate claim information based on the sensor data and may process the claim information based on at least one insurance profile maintained by the autonomous vehicle control system. Then, the autonomous vehicle control system may generate a claim notification based on processing the claim information and may send the claim notification to a vehicle management computer system.

Systems, methods, and non-transitory computer-readable media can determine sensor data captured by at least one sensor of a vehicle while navigating an environment over a period of time. Information describing one or more agents associated with the environment during the period of time can be determined based at least in part on the captured sensor data. A parameter-based encoding describing the one or more agents associated with the environment during the period of time can be generated based at least in part on the determined information and a scenario schema, wherein the parameter-based encoding provides a structured representation of the information describing the one or more agents associated with the environment. A scenario represented by the parameter-based encoding can be determined based at least in part on a cluster of parameter-based encodings to which the parameter-based encoding is assigned.

A physical space may contain sliding doors in various places such as between rooms, as closet doors, on furniture, and so forth. An autonomous mobile device (AMD) capable of non-holonomic motion may selectively contact a portion of the sliding door and apply a force to open or close the door. The AMD moves to an initial pose relative to the door, with a portion of the AMD coming into contact with the door. The AMD then rotates and translates, relative to the door, applying a force to the door. Sensor data, such as linear acceleration, rotation, drive wheel torque, and so forth is used as input to determine the next movement in a series of motions, resulting in continued contact and force application. As the end of travel for this motion is reached, the AMD may separate from the door, reposition, and continue to apply a force to the door.

A physical space may contain sliding doors in various places such as between rooms, as closet doors, on furniture, and so forth. An autonomous mobile device (AMD) capable of non-holonomic motion may selectively contact a portion of the sliding door and apply a force to open or close the door. The AMD moves to an initial pose relative to the door, with a portion of the AMD coming into contact with the door. The AMD then rotates and translates, relative to the door, applying a force to the door. Sensor data, such as linear acceleration, rotation, drive wheel torque, and so forth is used as input to determine the next movement in a series of motions, resulting in continued contact and force application. As the end of travel for this motion is reached, the AMD may separate from the door, reposition, and continue to apply a force to the door.

Navigation systems and methods for autonomous vehicles are provided. The navigation system may include multiple navigation subsystems, including one having an inertial measurement unit (IMU). That unit may serve as the primary unit for navigation purposes, with other navigation subsystems being treated as secondary. The other navigation subsystems may include global positioning system (GPS) sensors, and perception sensors. In some embodiments, the navigation system may include a first filter for the IMU sensor and separate filters for the other navigation subsystems.

A HAPS platform may execute a neural network (a HAPSNN) as it monitors air traffic; the neural network enables it to classify, predict, and resolve events in its airspace of coverage in real time as well as learn from new events that have never before been seen or detected. The HAPSNN-equipped HAPS platform may provide surveillance of nearly 100% of air traffic in its airspace of coverage, and the HAPSNN may process data received from a drone to facilitate safe and efficient drone operation within an airspace.

A HAPS platform may execute a neural network (a HAPSNN) as it monitors air traffic; the neural network enables it to classify, predict, and resolve events in its airspace of coverage in real time as well as learn from new events that have never before been seen or detected. The HAPSNN-equipped HAPS platform may provide surveillance of nearly 100% of air traffic in its airspace of coverage, and the HAPSNN may process data received from a drone to facilitate safe and efficient drone operation within an airspace.

A patrol robot including a plurality of magnetic sensors for detecting a magnetic marker laid on a traveling road has at least two or more magnetic sensors arrayed on a sensor array line linearly extending along any direction. In the patrol robot, two sensor array lines are formed, and since at least any one sensor array line can cross with respect to a relative moving direction of the magnetic marker with a movement of the patrol robot, the magnetic marker can be detected with high reliability, irrespective of the moving mode.

The present disclosure provides a robot control method, a robot, a control terminal, and a control system. The method includes: obtaining an environment feature around the first robot when the first robot detects no positioning identifier; determining a deviation distance and a deviation angle between the first robot and a target traveling route of the first robot according to the environment feature and traveling information of the first robot; and controlling the first robot to perform route correction according to the deviation distance and the deviation angle, to cause the first robot to move to the target traveling route again.

The present disclosure provides a robot control method, a robot, a control terminal, and a control system. The method includes: obtaining an environment feature around the first robot when the first robot detects no positioning identifier; determining a deviation distance and a deviation angle between the first robot and a target traveling route of the first robot according to the environment feature and traveling information of the first robot; and controlling the first robot to perform route correction according to the deviation distance and the deviation angle, to cause the first robot to move to the target traveling route again.