A computer generates historical time and velocity data for vehicles based on data from sensor(s) observing the vehicles. The computer determines, based on the historical time and velocity data, a control policy that controls movement of the vehicles. The control policy is represented as a weighted combination of a set of predefined policies. Determining the control policy comprises calculating weights or parameters for a weighted combination of the set of predefined policies that minimizes a residual error term. The residual error term is computed based on a difference between the historical time and velocity data and predicted time and velocity data associated with the weighted combination of the set of predefined policies. The computer determines an action plan based on the determined nonlinear control policy. The computer transmits, to a machine, a control signal causing the machine to perform or simulate at least a part of the determined action plan.

Systems and methods for targeted operational disablement that may utilize, for example, UAV-delivered shaped charges coupled to specific features of a target object. A UAV may navigate to, identify, and attach a disabling device to a critical component of a target such as a barrel of a tank main gun, a crew hatch cover, a tank tread, an engine compartment, or a magazine.

A re-usable intercept drone (104) comprises an elongate fuselage (200), a first wing (202) and a second wing (206) operably coupled to the elongate fuselage (200) and extending substantially away from the elongate fuselage (200). A first propulsion unit (210) and a second propulsion unit (212) are operably coupled to the first wing (202) and the second wing (206), respectively. A third propulsion unit (214) and a fourth propulsion unit (218) are operably coupled to the fuselage (200). The first, second, third and fourth propulsion units (210, 212, 214, 218) are circumferentially spaced about the elongate fuselage (200).

Systems, frameworks, and methods for reinforcement learning (RL)-based real-time path planning for unmanned aerial vehicles (UAVs) are provided. The learning capabilities of federated learning (FL) can be integrated with an improved deep RL framework, including using a significant reply memory buffer (SRMB) to accelerate the intelligent behavior. The framework can train a UAV to intelligently dodge static and dynamic defense systems and achieve assigned goals (e.g., in a hostile area). The FL can enable collaborative learning through a swarm of UAV agents.

The present disclosure provides vehicle collision avoidance system (CAS) that includes a first CAS module for a robotic vehicle that includes an interface to a first network; a hard stop interface communicatively coupled to the robotic vehicle; one or more computer processors; one or more computer readable storage media; and program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the stored program instructions including instructions to: determine a relative position of a second CAS module based on information from the first network; responsive to determining that the second CAS module is within a first distance from the first CAS module, issue an alert; and responsive to determining that the second CAS module is within a second distance from the first CAS module, transmit a hard stop signal to the robotic vehicle.

A mobile laser denial defense system is provided. A camera is in communication connection with an image processing device and a control device, and is configured for real-time monitoring and target identification; a laser is fixed on a pan/tilt head and is connected to a laser host by a telescopic optical fiber, where the laser is in communication connection with the control device; a cooling system is connected to the laser and is configured to cool and dissipate heat from the laser; a control device is connected to a camera and the laser and configured to identify a target and control the laser to perform laser strikes, where the control device is also connected to the cooling system to monitor a temperature state of the cooling system in real time; and an automatic cruise and obstacle avoidance system is integrated in a mobile base.

Unmanned systems, and primarily vehicles, and in most embodiments unmanned aerial systems (UAS), as well as novel guided and unguided munitions are presented herein. to the two combine to present unmanned weaponized systems that not only carry weapons functionality, but allow for advanced observation and reconnaissance. These unmanned systems are largely directed toward military applications where the weaponized unmanned system may be forward deployed to allow human warfighters to remain at long range distances from the potential targets, and to provide surveillance, target identification and tracking, general reconnaissance information, and the like, while also providing weaponized attack capabilities. The unmanned systems may, in some embodiments, include advanced command and control capabilities for communication between the system and the remote, rear-positioned warfighter, and between separate elements of the unmanned system. Many embodiments also include weapons systems, munitions, or rounds with guidance and real-time maneuverability capabilities as well.

A computerized system configured to execute a multi-agent machine learning model for controlling a plurality of vehicles in a multi-vehicle autonomous control session in a multi-vehicle environment is disclosed. Multi-modal neural network agents of the model each control a corresponding autonomous vehicle in the session. The agents receive image data and parameter data, input the image data to an image feature extractor to produce an image feature vector, input the parameter data to a parameter data feature extractor to produce a parameter data feature vector, produce a joint latent representation of the image data and parameter data, and input the joint latent representation to an actor model neural network, to generate a selected action for the autonomous vehicle. The multi-agent machine learning model is configured to control each autonomous vehicle in the session according to the corresponding selected action for each autonomous vehicle.

Disclosed is a boxing training system with a feedback function, including a training robot, a boxing glove assembly and a virtual reality device. The training robot performs a displacement movement. The boxing glove assembly is located on a user's hand and hits the training robot. The virtual reality device is electrically connected to the training robot and the boxing glove assembly, respectively. The virtual reality device is located at the user's eyes and displays a virtual image. The virtual image has a virtual opponent and a virtual boxing glove. When the virtual boxing glove hits the virtual opponent, the boxing glove assembly just hits the training robot.

An air-based counter-countermeasure (CCM) system includes an autonomous or semi-autonomous unmanned aircraft capable of deployment to an area of a countermeasure system having (1) a laser detector to receive and identify an incident laser signal as a target designator signal and (2) one or more countermeasure subsystems operative in response to an output of the laser detector to deploy corresponding countermeasures against a laser target designator. The CCM system further includes a laser-based CCM subsystem carried by the aircraft, the CCM subsystem being configured and operative, during the deployment of the aircraft, to direct a simulated target designator laser signal to the laser detector of the countermeasure system to trigger one or more of the countermeasures and thereby reduce protective ability of the countermeasure system against subsequent laser-guided attack.