Anti-drone firearm use training system

Abstract

An anti-drone firearm training system comprises an orientation detector attachable, and in use attached, to a pointer, such as a rifle, and apparatus to detect an action initiated by a user of the poi

Description

The present invention relates to an anti-drone firearm use training system. Background Technological miniaturization has caused the diversification of aerial vehicles in both size and purpose. Manned, remote controlled and autonomous vehicles simultaneously operate alongside traditional aeroplanes and helicopters with quadcopters and insectoid aerial vehicles all filling different roles. Unmanned aerial vehicles, drones, may be used as a part of undesirable activities, including military application by opposing forces, use by undesired criminal operations and accidental misuse. In these scenarios prompt termination of flight, by a third party, may be desired. Shooting the drone with a firearm can be a solution. However, drones, particular in the form of micro aerial vehicles, such as in the ten kilogram or lower weight class, presented a target which is unconventional for conventional firearms, being relatively high in the air and relatively small. Training for the shooting down of drones is not currently adequately catered for, since current systems are dedicated for training against similar conventional aerial targets such as wildfowl and these exhibit different behaviours than exhibited by drones. Even more so, firearms training related to persons and animals on land vehicles and aircraft are again a very different situation and hence prior training does not extrapolate and thus is not adequate. There is therefore a need for an anti-drone firearms training system to give actively train users, such as the police or military. The present invention The present invention in its various aspects is as set out in the appended claims. The present invention provides an anti-drone firearm training system comprising an orientation detector attachable, and in use attached, to a pointer and apparatus to detect an action initiated by a user of the pointer, means for ascertaining a location of a drone in-flight, and a controller in communication with the detector, the apparatus and the means; the controller being configured: a) to determine, upon the apparatus detecting the action initiated by a user; b) a location in space of the drone and the orientation of the pointer; c) by means of a trained computing system to determine, based upon a predetermined set of characteristics of the pointer, and upon a predetermined set of characteristics of the action initiated: d) to determine a probability that the action initiated would cause an actual or notional interaction with the drone in flight as determined by the computing system. The present invention brings together the best elements from similar nature training and practical training. By use of the physical pointer, which is preferably an actual firearm, with a physical drone, preferably operated out of doors then the physical impediments and limitations of the system are readily apparent and incorporated. However, unlike with target shooting on the ground where the grouping of shots can be measured, such as by use of a paper target or similar, this is not possible with a drone at elevation. Therefore, there is no constructive feedback to the trainee, i.e. the user, as to how close they may or may not have been to an accurate shot. In addition, the system of the present invention avoids the firing of live ammunition at elevation, which must perforce land somewhere, typically a long distance away and therefore be inherently dangerous. The computing system requires training using actual equipment, but once trained, or simply programs with relevant parameters can provide the user with feedback. In particular the determination of probability of an interaction, such as feedback audibly, visually or by other means provides efficient means for training. Whilst the invention is described in generalised terms regarding characteristics and probabilities and interactions the skilled person will appreciate that these are generalised terms to describe the ballistic parameters of a firearm, grouping of shots and the potential for an actual or notional shot to hit the drone. In summary, the present invention marries the best aspects of physical training with the safety and economy of a simulation. In the present invention the computing methodology may be a trained artificial neural network; the artificial neural network being trained using data of one or more firearms models, one or more ammunition specifications, one or more drone types; and a set of orientation parameters; a set of location parameters relative to the point of determination of the orientation parameters and a calculated probability of intersection of said ammunition coming into proximity with the drone in the given location relative to the firearm upon discharge of the ammunition from the firearm at the given orientation based upon the training. The training comprising back-propagation in which the probability of intersection in space is corrected in order to make the results coincide with the training objectives of a positive correlation wherein the back-propagation is performed via an algorithm provided by a human and the correction is achieved by the artificial neural network adjusting its trained dataset by adjusting weighting and bias in nodes and levels of assessment of an intersection in the neural network. The computing methodology in use attributing a model to the pointer, the specification to one such specification selected by the user of the system; the type corresponding to one of such type selected by the user of the system. In the present a record of actual physical interactions between an emission, such as ammunition, from a firearm and a drone in flight may be used to train the computing methodology. In the present invention the training data may include an effect on the likelihood of intersection of an interval between the actions initiated in that time interval may be used as one of the characteristics of the action initiated. In the present invention the predetermined set of characteristics of the action being that of ammunition may comprise a plurality of projectiles in a single discharge. In the present invention the means for ascertaining a location the location of the drone in-flight may comprise a GPS location finder located upon the drone configured to communicate that information wirelessly to the controller in real time. In the present invention the means for ascertaining a location the location of the drone in-flight may comprise a camera in a known and fixed location relative to or being part of part of the orientation detector, the camera being configured to communicate a position in the field of view of the camera to the controller. In the present invention the orientation detector may comprise a GPS location finder and a three-axis accelerometer. In the present invention the controller uses information from a return time-of-flight of a signal between the controller and / or the orientation detector to supplement the determination by the controller of the location of the drone relative to the orientation detector. In the present invention the system, in use, may comprise a firearm or a simulated firearm as the pointer. In the present invention the orientation detector may be attached to the picatinny rail of the firearm or simulated firearm. In the present invention the apparatus to detect an action initiated by a user may comprise a switch attached to the pointer and actuated by a user of the system to determine said action. In the present invention the apparatus detecting action initiated by a user may comprise an accelerometer configured to detect axial movement along an elongate axis of the pointer. The present invention includes an artificial neural network trained to carry out the computational actions of the system of any preceding claim. The present invention includes a kit of parts for use with a pointer and a drone to construct the system of the present invention, the kit of parts including the orientation detector and the controller of any preceding claim. The principal benefit of the present invention is that it enables the use of specific, actual, equipment, a drone, a firearm and a relevant environment for training whilst giving feedback in terms of a probability of success. This enables the user to try various techniques, methodologies and procedures and evaluate their effectiveness. In particular it does so without causing any destruction. Using an actual firearm with live ammunition gives negligible feedback for anything other than a hit and a hit, which may simply be a lucky circumstance would destroy part of the equipment. In a further embodiment of the present invention the apparatus is configured to use an actual firearm with live ammunition but in which upon detecting the action of the user, i.e. pulling the trigger signals to the drone to change its position, preferably in conjunction with providing a predicted intersection so that a calculation can be carried out to avoid the projectile / ammunition. In other words, in the further embodiment the equipment works both forwards and backwards, i.e. the potential for a shot to be on target is evaluated but also, when real ammunition is used target is prewarmed and moves out of the way. Whilst this latter action is not a valid simulation in itself the feedback of a high probability to hit provides the constructive feedback for the training whilst the avoidance by the target preserves the equipment. The present invention may comprise a system comprising computer system and camera suitable for training anti-drone firearm use. The camera as an input into the computer system. Preferably the camera being a visible light colour digital video camera. The computer system computing the probability of hitting the target. Preferably including expected number of hits and expected effectiveness thereof. Probability being particularly relevant for shotgun and other firearm which’s shot patterns are expected to have a high degree of randomness. Preferably the computer system being Al. Preferably the computer system being able to calculate hit probabilities factoring in: firearm and cartridge properties. Preferably able to compare possible cartridges for a given application. Preferably the computer system able to feedback to the user or training supervisor for training and / or testing purposes. The system preferably further comprising component suitable to be mounted to a firearm. Preferably the attachment not distorting the operation of the firearm for training purposes. Preferably the component being mounted on the firearm alongside other attachments, which may include optics, without hinderance. Preferably the component suitable to be mounted to a firearm can be unmounted from the firearm. The system preferably suitable to be interfaced to a firearm with a blank-firing adaptor. The system preferably where the component suitable to be mounted comprises a camera. Preferable the camera facing in the same direction as the barrel of the firearm. The system preferably comprising two or more cameras. Preferably the multiple images from multiple perspectives. The system preferably comprising three or more cameras. The system preferably comprising a drone. The drone to be a target. The system preferably comprising an accelerometer. Preferably an accelerometer on the / each camera. Preferably an accelerometer on the component suitable to be mounted to a firearm. Preferably an accelerometer on the drone. The system preferably comprising two or more accelerometers. Preferably the accelerometers having different ranges of sensitivity. The system preferably comprising a gyroscope. Preferably a gyroscope on the / each camera. Preferably a gyroscope on the component suitable to be mounted to a firearm. Preferably a gyroscope on the drone. The system preferably comprising GPS. Preferably GPS on the / each camera. Preferably GPS on the component suitable to be mounted to a firearm. Preferably GPS on the drone. The system preferably comprising wind measurement system. Preferably measuring wind direction. Preferably measuring wind speed. The computer system preferably calculating the wind effect on probable projectile trajectory. The system preferably with the firing of the firearm input into the computer system. Preferably the computer system calculating probability of hit of shot fired at time of trigger pull, calculating probability of hit of shot fired in a time interval after trigger pull as if the shooter had delayed, probability of hit of shot fired in time interval before trigger pull, as if the shooter had hastened. A kit of parts for the above. The term ‘firearm’ may be extrapolated to include functionally equivalent projectile firing technologies. An alternative embodiment of the present invention the calculation is carried out using Euclidean geometry in conjunction with statistics to determine the probability of an interaction. I.e. an Al solution is not present.

Claims

1. An anti-drone firearm training system comprising an orientation detector attachable, and in use attached, to a pointer and apparatus to detect an action initiated by a user of the pointer, means for ascertaining a location the location of a drone in-flight, and a controller in communication with the detector, the apparatus and the means; the controller being configured:a) to determine upon the apparatus detecting the action initiated by a user;b) a location in space of the drone and the orientation of the pointerc) by means of a trained computing methodology to determine based upon a predetermined set of characteristics of the pointer and upon a predetermined set of characteristics of the action initiated:d) to determine a probability that the action initiated would cause an interaction with the drone in flight as determined by the computer methodology.

2. The system of claim 1 wherein the computing methodology is a trained artificial neural network; the artificial neural network being trained using data of one or more firearms models,one or more ammunition specifications, one or more drone types;a set of orientation parameters;a set of location parameters relative to the point of determination of the orientation parameters and a calculated probability of intersection of said ammunition coming into proximity with the drone in the given location relative to the firearm upon discharge of the ammunition from the firearm at the given orientation the training comprising;the training comprising back-propagation in which the probability of intersection in space is corrected in order to make the results coincide with the training objectives of a positive correlation wherein the back-propagation is performed via an algorithm provided by a human and the correction is achieved by the artificial neural network adjusting its own internal workings in such A way as toadjusting weighting and bias in its nodes and levels of assessment of an intersection;the computing methodology in use attributing a model to the pointer, the specification to one such specification selected by the user of the system; the type corresponding to one of such type selected by the user of the system.

3. The system of claim 1 or claim 2 wherein a record of actual physical interactions between an emission, such as ammunition, from a firearm and a drone in flight is also used to train the computing methodology.

4. The system of claim 2 or claim 3 were in the training data includes an effect on the likelihood of intersection of an interval between the actions initiated in that time interval is used as one of the characteristics of the action initiated.

5. The system of any preceding claim wherein the predetermined set of characteristics of the action being that of ammunition comprises a plurality of projectiles in a single discharge.

6. The system of any preceding claim wherein the means for ascertaining a location the location of the drone in-flight comprises a GPS location finder located upon the drone configured to communicate that information wirelessly to the controller in real time.

7. The system of any preceding claim wherein the means for ascertaining a location the location of the drone in-flight comprises a camera in a known and fixed location relative to or being part of part of the orientation detector, the camera being configured to communicate a position in the field of view of the camera to the controller.

8. The system of any preceding claim wherein the orientation detector comprises a GPS location finder and a three-axis accelerometer.

9. The system of any preceding claim wherein the controller uses information from a return time-of-flight of a signal between the controller and / or theorientation detector to supplement the determination by the controller of the location of the drone relative to the orientation detector.

10. The system of any preceding claim wherein the system, in use, comprises a firearm or a simulated firearm as the pointer.

11. The system of claim 10 wherein the orientation detector is attached to the picatinny rail of the firearm or simulated firearm.

12. The system of any preceding claim wherein the apparatus to detect an action initiated by a user comprises a switch attached to the pointer and actuated by a user of the system to determine said action.

13. The system of any preceding claim wherein the apparatus detecting action initiated by a user comprises an accelerometer configured to detect axial movement along an elongate axis of the pointer.

14. The artificial neural network of the system of any preceding claim.

15. A kit of parts for use with a pointer and a drone to construct the system of the present invention, the kit of parts including the orientation detector and the controller of any preceding claim.10