A management system includes a course data generation unit that generates course data for each of a plurality of unmanned vehicles such that loading work for the plurality of unmanned vehicles by a loader is sequentially performed on a work site where a plurality of the loaders operates; and a priority determination unit that determine a passage order at an intersection on the work site of the plurality of unmanned vehicles traveling according to the course data so as to reduce a total loading loss indicating a total of loss amounts in operation of each of the plurality of the loaders.

The present technology relates to a control device, a control method, and a program that make it possible for objects that are capable of moving autonomously to move in corporation with each other. The control device according to one aspect of the present technology moves, in response to an action made by a user on a predetermined object among multiple objects that are capable of moving autonomously, an object corresponding with the predetermined object. The present technology can be applied to a control device that controls mobile robots.

A movable body control system (SYS) includes first and second movable bodies (1#1, 1#2) that are movable in a predetermined area (TA) in which a wireless communication network (NW) is built; and a control apparatus (3) for controlling the first and second movable bodies through the wireless communication network, the control apparatus includes: a storage unit (32) for storing a first communication quality information that indicates a communication quality in the predetermined area when the first movable body exists in each of a plurality of different locations in the predetermined area; a generation unit (311) for generating, based on the first communication quality information, a target moving route (TGT#2) that allows the second movable body to move while avoiding a first low quality location (DA_low1) at which the communication quality does not reach a desired quality due to the first movable body; and a control unit (312) for controlling the second movable body so that the second movable body moves along the target moving route.

A combinable and detachable vehicle includes: a front connecting portion and a rear connecting portion provided at a front portion and a rear portion of a vehicle, respectively, and connected to another vehicle when the vehicle is combined with the other vehicle; a front monitoring portion and a rear monitoring portion monitoring a forward area and a rearward area of the vehicle, respectively; and a controller determining whether the other vehicle is connected to the front connecting portion or the rear connecting portion of the vehicle based on information received from the front connecting portion or the rear connecting portion, or the front monitoring portion or the rear monitoring portion, and controlling whether to operate each of a front component and a rear component of the vehicle depending on a direction of the vehicle is connected to the other vehicle.

In various examples, a sequential deep neural network (DNN) may be trained using ground truth data generated by correlating (e.g., by cross-sensor fusion) sensor data with image data representative of a sequences of images. In deployment, the sequential DNN may leverage the sensor correlation to compute various predictions using image data alone. The predictions may include velocities, in world space, of objects in fields of view of an ego-vehicle, current and future locations of the objects in image space, and/or a time-to-collision (TTC) between the objects and the ego-vehicle. These predictions may be used as part of a perception system for understanding and reacting to a current physical environment of the ego-vehicle.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for predicting how likely it is that a target agent in an environment will yield to another agent when the pair of agents are predicted to have overlapping future paths. In one aspect, a method comprises obtaining a first trajectory prediction specifying a predicted future path for a target agent in an environment; obtaining a second trajectory prediction specifying a predicted future path for another agent in the environment; determining that, at an overlapping region, the predicted future path for the target agent overlaps with the predicted future path for the other agent; and in response: providing as input to a machine learning model respective features for the target agent and the other agent; and obtaining the likelihood score as output from the machine learning model.

Aspects of the disclosure relate generally to controlling an autonomous vehicle in a variety of unique circumstances. These include adapting control strategies of the vehicle based on discrepancies between map data and sensor data obtained by the vehicle. These further include adapting position and routing strategies for the vehicle based on changes in the environment and traffic conditions. Other aspects of the disclosure relate to using vehicular sensor data to update hazard information on a centralized map database. Other aspects of the disclosure relate to using sensors independent of the vehicle to compensate for blind spots in the field of view of the vehicular sensors. Other aspects of the disclosure involve communication with other vehicles to indicate that the autonomous vehicle is not under human control, or to give signals to other vehicles about the intended behavior of the autonomous vehicle.

Apparatus and methods related to routing robots are provided. A roadmap of an environment that includes first and second robots can be received. The roadmap can be annotated with unidirectional lanes connecting conflict regions, where each lane ends so to avoid blocking a conflict region. First and second routes for the respective uses of the first and second robots can be determined, where both the first and second routes include a first lane connected to a first conflict region. A first, higher priority and a second, lower priority can be assigned to the respective first and second robots. It can be determined that the second robot following the second route will block the first robot on the first lane. Based on the first priority being higher than the second priority, the computing device can alter the second route to prevent the second robot from blocking the first robot.

Aspects of the disclosure relate to testing predictions of an autonomous vehicle relating to another vehicle or object in a roadway. For instance, one or more processors may plan to maneuver the autonomous vehicle to complete an action and predict that the other vehicle will take a responsive action. The autonomous vehicle is then maneuvered towards completing the action in a way that would allow the autonomous vehicle to cancel completing the action without causing a collision between the first vehicle and the second vehicle, and in order to indicate to the second vehicle or a driver of the second vehicle that the first vehicle is attempting to complete the action. Thereafter, when the first vehicle is determined to be able to take the action, the action is completed by controlling the first vehicle autonomously using the determination of whether the second vehicle begins to take the particular responsive action.

Aspects of the technology relate to exception handling for a vehicle. For instance, a current trajectory for the vehicle and sensor data corresponding to one or more objects may be received. Based on the received sensor data, projected trajectories of the one or more objects may be determined. Potential collisions with the one or more objects may be determined based on the projected trajectories and the current trajectory. One of the potential collisions that is earliest in time may be identified. Based on the one of the potential collisions, a safety-time-horizon (STH) may be identified. When a runtime exception occurs, before performing a precautionary maneuver to avoid a collision, waiting no longer than the STH for the runtime exception to resolve.