A method is provided of real-time controlling a remote device to perform a task, the method comprising steps of: for controlling the remote device to perform a task, obtaining graphical data, such as image frames forming a video, of surroundings of the remote device, such as an area of farmland or beach, sending the graphical data to a remote operation device, obtaining user input data from an operator, which user input data is indicative of a location of interest in the graphical data, generating a control signal for controlling the remote device to perform a task based on the user input data, and using the control signal for controlling the remote device to perform the task at the location of interest. The user input data is further used as training data for training a machine learning algorithm, which algorithm is arranged for generating at least part of a control signal for controlling the remote device; and/or providing a suggested location of interest to the operator.

An automatic wayfinding system is provided and includes a robotic guide, a passenger interface through which a passenger provides an input of a desired destination and a dispatching module, which, upon receipt of the input of the desired destination, dispatches the robotic guide to the passenger. At least one of the robotic guide and the dispatching module have access to a navigation map including the desired destination, a current position of the passenger and information relating to one or more paths from the current position of the passenger to the desired destination. The robotic guide is programmed to lead the passenger along the one or more paths while remaining tethered to the passenger.

Aspects relate to systems and methods for guided autonomous locomotion of a carriage, including a compartment configured to ensconce a child, a frame configured to support the compartment, a drive motor, a drivetrain operatively coupled to the drive motor; a drive wheel rotatably affixed to the frame, configured to contact a support surface and operatively coupled to the drivetrain, wherein operating the at least a drive motor causes the at least a drive wheel to rotate, an environmental sensor configured to sense an environmental characteristic related to an environment substantially surrounding the carriage; a battery configured to power the at drive motor and a controller configured to control the drive motor in response to the environmental characteristic.

A golf vehicle following system includes a following golf vehicle. The following golf vehicle includes a driveline, a communications interface, and a sensor system configured to acquire second data. The golf vehicle system includes at least one processing circuit having at least one processor and at least one memory, the at least one memory storing instructions thereon that, when executed by the at least one processor, cause the at least one processor to: receive a request for the following golf vehicle to follow a leading golf vehicle; and control the driveline such that the following golf vehicle follows the leading golf vehicle within a specified distance based on at least one of (a) first data acquired from at least one of the communications interface or from a global positioning system or (b) the second data acquired by the sensor system.

Method, apparatuses and computer program products for training machine learning models for logistics robots and routing the robots are disclosed. A method of training an ML model involves providing a route to a logistics robot, obtaining route issue indications from the robot when it fails to traverse the route as expected, and associating map objects with the issue locations. The trained model can then be used to determine the likelihood of route issues for a specific logistics robot type based on the presence of certain map objects along the route. The disclosure further involves calculating route penalty value(s) based on the likelihood and updating the route accordingly, including selection of a logistics robot type for the route.

A modular micro-factory swarm for autonomous urban construction is disclosed. Each unit is a self-assembling robotic tile that (i) 3D-prints structural elements from locally characterized feedstock, (ii) powers itself via a dual renewable system (deployable photovoltaic array and hydrogen electrolysis/fuel-cell), and (iii) communicates over a mesh network for coordinated tasking. A cost-based planner assigns paths and jobs to minimize energy, time, and terrain risk. After each extrusion, the tile runs on-board structural checks (ultrasonic echo and vibrational resonance) and records pass/fail metrics. Build provenance is bound cryptographically: a hardware security module signs a build object identifier that hashes tile ID, task node, material signature, time, and location; signed records are committed to a distributed ledger. Actuation is gated by location-specific consent tokens verified against policy maps before printing proceeds. The architecture enables peer-to-peer orchestration, verifiable quality control, and closed-loop energy autonomy, reducing reliance on infrastructure and supervision.