Systems and method are provided for predicting a vehicle pose. In one embodiment, a system includes a non-transitory computer readable medium. The non-transitory computer readable medium includes: an asynchronous prediction module configured to, by a processor, receive vehicle data from a vehicle, and compute a prediction of vehicle pose from the vehicle data, wherein the computation is performed asynchronously using a vehicle time frame and a common coordinate system; and a synchronous prediction module configured to, by the processor, compute a final prediction of the vehicle pose across the vehicle population based on the prediction of vehicle pose from the asynchronous prediction module, wherein the computation is performed synchronously using a global time frame and the common coordinate system.

A method navigates a robotic mower (2) comprising at least one sensor (12; 14). The method comprises detecting (S101), by means of the at least one sensor (12, 14), at least one signal from a wire (8) as a first signal source; controlling (S106) the robotic mower (2) to follow the wire (8) at a first distance (D1) to the wire (8); detecting (S108), by means of the at least one sensor (12, 14), at least one signal from a second signal source; detecting (S109) a second distance (D2) based on the at least one signal from the second signal source; and navigating (S110) the robotic mower (2) along a path determined based on the first distance (D1) and the second distance (D2).

A software product and methods determine a field coverage method for a nonholonomic robot to process a field using parallel lanes. A cellular decomposition algorithm divides the field into a plurality of cells, each having a plurality of parallel lanes. Permutations of lane processing orders are determined for each cell, based upon a minimum turning radius of the robot. A cell graph is generated to determine a shortest path for single-time processing each lane in each cell without violating the minimum turning radius of the robot. A step list defining movement of the nonholonomic robot along each lane in each cell of the shortest path through the cell graph is generated, and transits between the lanes, and laps around the field and any obstacles are added. A path program to control the nonholonomic robot to process the field is generated based upon the step list.

A serving system using a robot may include a caller mounted on each of a plurality of tables, and a robot driving to a target table among the plurality of tables based on information received from the caller. The robot may store information on the location of each of the plurality of callers, the size and shape of each of the plurality of tables, the location of the caller mounted on the table, and the rotational angle with respect to a baseline of the caller. The caller may transmit to the robot information on the distance from the caller to the robot and the direction of the robot with respect to the caller during driving of the robot. The robot may transmit and receive a wireless signal over a mobile communication network constructed according to 5G (Generation) communication.

An autonomous skid-steer machine includes a chassis, a plurality of ground engaging elements supporting the chassis on a ground surface and one or more computing devices for controlling movement of the machine. The one or more computing devices are configured to generate a control signal for controlling operation of at least some of the plurality of ground engaging elements, the control signal being generated according to a control algorithm incorporating tuning parameters, and to automatically determine the tuning parameters in real time during operation of the machine such that the control algorithm is always optimized for a current speed of the machine. The tuning parameters are found using the machine's current speed, a natural frequency value and a damping value.

According to an embodiment, a system generates a number of sample trajectories from a trajectory sample space for a driving scenario. The system determines a reward based on a reward model for each of the sample trajectories, where the reward model is generated using a rank based conditional inverse reinforcement learning algorithm. The system ranks the sample trajectories based on the determined rewards. The system determines a highest ranked trajectory based on the ranking. The system selects the highest ranked trajectory to control the ADV autonomously according to the highest ranked trajectory.

A travel route determination system including: a route generating unit generating planned travel routes including work routes along which a work vehicle performs autonomous travel; a control unit capable of causing the work vehicle to perform autonomous travel along the planned travel routes; an information obtaining unit obtaining position information and orientation information on the work vehicle; and a determination unit determining an autonomous travel candidate route at which the work vehicle can start autonomous travel, before the work vehicle starts autonomous travel. The determination unit sets a candidate determination region based on the position information and orientation information on the work vehicle, and the determination unit determines, among the work routes, a work route in the candidate determination region as the autonomous travel candidate route.

A control system for a work vehicle includes a storage device and a controller in communication with the storage device. The storage device stores target parameter data defining a relationship between a movement distance of the work vehicle and a target parameter related to a target digging amount of a work implement of the work vehicle. The target parameter data includes digging time data defining a relationship between the movement distance of the work vehicle within a predetermined digging area and the target parameter. The controller determines a target return distance from a distance of the digging area defined in the target parameter data, and determines a position returned from a predetermined reference position by the target return distance as a recommended digging start position.

A method for performing a pickup/drop-off at a location includes providing, to an autonomous vehicle (AV), a parameter value of a parameter associated with the pickup/drop-off. The AV executes the pickup/drop-off according to the parameter value. After the pickup/drop-off is executed, feedback related to the parameter is received from a customer and from a tele-operator. The method also includes identifying other world objects present at the location during the pickup/drop-off and generating an optimized parameter value for executing the pickup/drop-off using the parameter value, the customer feedback, the tele-operator feedback, and other world objects.

A traveling vehicle system includes a plurality of traveling vehicles; and a host controller in which: a first deriver derives a first traveling vehicle scheduled to pass through a branching part through which a priority traveling vehicle selected by a priority traveling vehicle selector is scheduled to pass; a merging part deriver derives a merging part through which the first traveling vehicle is scheduled to pass after the branching part; and passage permission for the merging part is transmitted to the first traveling vehicle in priority to a second traveling vehicle scheduled to pass through the merging part from a direction different from a direction from which the first traveling vehicle is scheduled to pass through the merging part.