WO2022101637A1 - Improvements in or relating to vehicle safety in a dynamic environment - Google Patents

Abstract

A method for improving the safety of remotely controlled vehicles in a dynamic environment is provided. The method comprising the steps of: accessing a baseline model of an environment; monitoring at least one dynamic feature within the environment; modifying the baseline model to incorporate the at least one dynamic feature being monitored; determining whether the modified model meets a predetermined set of criteria; and sending an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

Description

IMPROVEMENTS IN OR RELATING TO VEHICLE SAFETY IN A DYNAMIC ENVIRONMENT

FIELD OF THE INVENTION

The present invention relates to improvements in or relating to a method for improving the safety of remotely controlled vehicles in a dynamic environment and, more specifically, to monitoring dynamic features to identify when predetermined criteria indicating safe passage are met.

BACKGROUND TO THE INVENTION

The growing use of devices that transmit radio waves throughout our civilisation presents an opportunity to make use of these signals as a diagnostic tool to assess the stability of our society, cities, infrastructure and natural environment, which allow us to detect, in near real-time, when unexpected changes are beginning to occur or have actually taken place. Such a capability is dependent upon the observation, analysis and recording of the radio emissions and their scattering by manmade and natural phenomena over lengths of time that corresponds to the patterns of life. These patterns repeat on daily, weekly, yearly and other periods and unexpected events may disrupt these patterns. Early warning of these unexpected changes would permit stake holders in the efficient running and safe conduct of human society to act in a timely way to alleviate many adverse effects of the unexpected events causing the changes, or to monitor progress in planned or desired changes.

It is against this background that the present invention has arisen.

SUMMARY OF THE INVENTION

According to the present invention there is provided a method for improving the safety of remotely controlled vehicles in a dynamic environment, the method comprising: accessing a baseline model of an environment; monitoring at least one dynamic feature within the environment; modifying the baseline model to incorporate the at least one dynamic feature being monitored; determining whether the modified model meets a predetermined set of criteria; and sending an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

An output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met prevents the vehicle from entering a hazardous environment, which, in turn, reduces the likelihood of a crash or accident occurring. The remote controlled vehicles may be autonomous, semi-autonomous or radio controlled. They may be drones.

The method may further comprise the step of generating a baseline model of an environment.

The baseline model for a given environment can be informed by observing habitual changes to the environment which undergo a daily, weekly, monthly and/or seasonal cycle. The baseline model of the environment will take into account the presence of immovable objects, such as buildings, but further information about their status may be augmented on the basis of cyclic changes. For example, a tree may be present in the environment, but its effect on the environment will differ considerably between the summer when it is fully leaf covered and the winter when it is not. As a further example, a building present in the environment will have walls at fixed locations, but the light reflection from them will depend on the time at which the sun rises, which obviously varies from day to day in an annual cycle.

The step of monitoring at least one dynamic feature within the environment may comprise monitoring at least one of: airflow; water courses; electromagnetic signal strength; light conditions; precipitation; and the location of moveable objects.

In some embodiments, a plurality of the aforementioned dynamic features may be monitored. For example, all of the aforementioned dynamic features may be monitored.

Monitoring airflow within the environment enables the wind speeds and/or wind direction to be modelled. In addition, aerodynamic effects around objects, such as buildings, trees and/or cars, can be modelled. This may be used to prevent the vehicle from entering into turbulent and/or extreme airflow conditions.

Monitoring water courses, both natural and man-made water courses can be important because they will reflect electromagnetic signals and therefore change the overall detected signal strength by means of multipath interference. This interference may be constructive or destructive. The reflectivity of the water's surface will depend on disruptions to the surface caused by ripples that arise either from turbulent flow as a result of the volume of water or from the wind. Furthermore, there may be floating objects carried on the water which will also contribute to changes in the reflectivity of the water to the electromagnetic signals.

Monitoring the electromagnetic signal strength enables areas of high, low and no signal strength to be identified and modelled, which may be used to reduce the likelihood of a remotely controlled vehicle from losing signal altogether. This may be achieved by monitoring the location of at least one electromagnetic transmitter; at least one electromagnetic sensor and/or at least one electromagnetically reflective surface such as a building wall or lake surface.

Monitoring light conditions enables the reflectivity of surfaces to be modelled, which, in turn, enables the model to more accurately account for potential variations in electromagnetic signal strength throughout the environment. A combination of meteorological effects, such as sunlight and wind will impact the drying of surface water and therefore changes in the reflectivity of the surface to electromagnetic signals

Monitoring precipitation conditions again enables the reflectivity of surfaces to be modelled, which, in turn, enables the model to more accurately account for potential variations in electromagnetic signal strength throughout the environment.

Finally, monitoring the location of movable objects, such as other vehicles, not only enables the physical location thereof to be modelled, but also enables the electromagnetic signal strength within the environment to be more accurately modelled by accounting for the change in reflectivity within the environment. For example, rapidly moving vehicles, such as a stream of fast traffic may cause a change of frequency, by means of the Doppler effect, of the electromagnetic signals that reflect from the surfaces of movable objects. This, in turn, can affect signal strengths in surrounding areas over a broader waveband than if no motion was present.

The other vehicles may be drones or other autonomous or semi-autonomous vehicles. Collision avoidance is an important aspect of the monitored data as other vehicles in the environment must be tracked and their trajectory predicted and/or followed in order to avoid collision.

The baseline model may comprise the location of at least one static feature. Modelling the location of static features, such as a building, enables the predetermined set of criteria to account for these objects. For example, the predetermined set of criteria may be configured to prevent vehicles from colliding with these static objects.

The baseline model may be modified in real-time or near real-time. Real-time or near real-time means a delay that is comparable with the response time of the remotely controlled vehicle. The cycle of measurement, computation and outputting instruction to the remotely operated vehicle must happen within a time sufficiently short that the vehicle has time to respond to the present circumstances, whilst they remain present. Near real-time in this context may therefore be understood to be a matter of a few minutes, for example 1, 2, 3 or 5 minutes. Near real-time may be less than 1 minute, for example less than 10 seconds, preferably less than 1 second. Modifying the model in real-time or near real-time ensures that the output is sent within a suitable timeframe such that the monitored dynamic features have not substantially changed.

The output may be configured to: allow the vehicle to enter the environment only when the predetermined set of criteria has been met; or at least one of: hold the vehicle in its current position when the predetermined set of criteria has not been met; and propose an alternative route for the vehicle when the predetermined set of criteria has not been met.

Generating one of the three aforementioned outputs enables the vehicle to be provided with the most suitable instructions given the environment. More specifically, proposing an alternative route may comprise changing altitude. In some embodiments, the alternative route may comprise instructing an airborne vehicle to land.

The environment may be a predetermined distance away from the vehicle. For example, the environment may be at least 10 meters away from the vehicle. However, any predetermined distance may be used, such as 50 meters, 25 meters, 10 meters, 5 meters, 1 meter or less than 1 meter. Once the vehicle is located within the predetermined distance of the environment, the predetermined criteria may be adjusted. The adjusted predetermined criteria may have a lower threshold than the initial predetermined criteria such that the vehicle is only re-routed in an emergency situation.

The method may further comprise: predicting at least one dynamic feature based on previously monitored dynamic features stored within a memory and/or external data; and modifying the baseline model to incorporate the at least one predicted dynamic feature.

Predicting at least one dynamic feature within the environment enables the output to account for features which cannot be monitored and/or features that are likely to change in the future. For example, the location of moveable objections, such as cars or other drones, (i.e. traffic conditions) may initially be predicted based on daily or weekly cycles. These predictions may be accounted for within the output, which may prevent the passage of a vehicle into a first environment if a subsequent environment comprises more suitable conditions. In this example, the predetermined set of criteria for the first environment may not be met due to the subsequent environment being preferable. However, it will be appreciated that any dynamic feature may be predicted and accounted for in the output.

The predicted dynamic feature and the monitored dynamic feature may be the same dynamic feature. Alternatively, the predicted dynamic feature and the monitored dynamic feature may be different dynamic features. This may enable the model to include predictions of dynamic features that are unable to be modelled. For example, if a specific monitoring device is broken. The predicted features therefore provide a level of redundancy.

The predicted dynamic feature may be a dynamic feature expected to occur within the environment in the future.

The ability to predict dynamic features enables a large environment to be modelled. For example, the predicted feature may be a feature that is present or is expected to be present within the environment but is unable to be monitored at the time that the method is being carried out. These features may include light conditions around sunrise and/or sunset.

The method may further comprise updating the predicted dynamic features based on monitored dynamic features. Updating the predicated dynamic features based on the monitored dynamic features enables future predictions for subsequent methods to be more accurate.

The method may further comprise: updating the baseline model to incorporate a subsequent environment; predicting at least one dynamic feature within the subsequent environment that is expected to occur when the vehicle is located within the subsequent environment; modifying the updated model to incorporate the at least one predicted dynamic feature within the subsequent environment; determining whether the modified updated model meets a predetermined set of criteria; and sending an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

Predicating a dynamic feature within a subsequent environment enables a more suitable route for the vehicle to be determined, which accounts for both detected present conditions and predicted future conditions.

Furthermore, according to the present invention there is provided a system for carrying out the method according to any preceding claim, wherein the system comprises: a memory comprising the base line model of the environment, a sensor configured to detect the electromagnetic waves emitted within the environment; a first processor configured to determine at least one parameter of at least one dynamic feature within the environment based on the electromagnetic waves detected by the sensor; a second processor configured to modify the baseline model such that it incorporates the at least one parameter of at least one dynamic feature; a third processor configured to determine whether the modified model meets a predetermined set of criteria; and a communication module configured to generate an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

A system configured to allow a vehicle to enter an environment only when a predetermined set of criteria has been met may be used to prevent the vehicle from entering a hazardous environment. This may significantly reduce the likelihood of a crash or accident occurring.

The system may further comprise a transmitter configured to emit electromagnetic waves within the environment.

Transmitters within the system may be used to more accurately obtain electromagnetic signal strengths within the environment. The transmitter may transmit electromagnetic waves, such as radio waves, within the environment. The transmitted waves may be detected by the sensors.

The first, second and third processors may be configured to operate continuously in real-time or near real-time. Processors configured to operate in real-time or near real-time enables the communication module to send the output within a suitable timeframe such that the monitored dynamic features have not substantially changed.

The sensor may be configured to detect the electromagnetic waves emitted within the environment is ground based.

The provision of ground based radio frequency monitoring devices avoids a requirement for access to buildings, masts etc. that would be required to obtain non-ground based sensed data.

Alternatively or additionally, the sensor may be configured to detect the wind speed and direction within the environment.

The invention will now be further and more particularly described, by way of example only, with reference to the accompanying drawings

FIGURES

Figure 1 is a schematic of the constituent parts of a system that may be used to put the present invention into effect;

Figure 2 is a schematic of a city scape that may be divided into one or more environments;

Figures 3A to 3C show, schematically, different paths taken by transmitted electromagnetic signals depending on features present within the environment;

Figure 4 shows, schematically, different paths taken by radio waves emitted from a transmitter in an environment including a water feature; Figure 5 shows, schematically, the different paths taken by radio waves in the presence of a natural phenomenon such as a tornado;

Figure 6 shows, schematically, the different paths taken by radio waves in the presence of a natural phenomenon such as an earthquake;

Figure 7 shows, schematically, the different paths taken by radio waves in the presence of a natural phenomenon such as a thunderstorm;

Figures 8A to 8D show, schematically, different paths taken when the sensor is also an emitter;

Figures 9A to 9C show, schematically, further examples including a moving transmitter in an aircraft passing through the environment; a spoofing transmitter intending to jam the legitimate transmissions of a satellite transmitter; and an environment in receipt of multiple ground based sources;

Figure 10 shows the top level integers that make up a system configured to execute the method according to the present invention;

Figure 11 shows the functionality of the individual RF sensors;

Figure 12 shows the sensor processing function in detail providing further detail on feature 59 of Figure 10;

Figure 13 shows the steps of the advanced signal processing;

Figure 14 shows a block diagram of the key aspects of the scattering centre, image correlation and anomaly detection;

Figure 15 shows a block diagram of the baseline environment model generation and maintenance; and

Figures 16A to 16E show the dynamic fluctuation in electromagnetic signal at a single location with different pointings.

DETAILED DESCRIPTION

The present invention relates to a method and system, and variations thereof, is a Radio Frequency (RF) Environmental Sensor System, herein after referred to as 'the Sensor', is a collection of radio detection sensors and radio transmitters that are either collected into a single housing or can be distributed over the complex built environment in single receivers/transmitters pairs or plurality of such device houses. The devices ae connected by a data communication system, e.g. the Internet, or plurality of data communication systems, so as to exchange data from the Sensor or plurality of such sensors, to a single point or multiple collaborating points of coordinated assessment, or Sensor Control System. The data transferred between the Sensor and the Sensor Control system includes but is not limited to the overall radiated RF power spectrum and its associated temporal variations. Other aspects of the radio waves would also be sensed including, but not limited to, polarisation, phase and modulated power waveform.

In British patent application GB2014801.1 a method and system are described that makes use of the radiated electromagnetic power from radio devices in the built environment for the purpose of assessing the risks to flying drones associated with the effects of multipath reflections of radio waves as well as from the effects of wind flow around buildings. The method described in that patent employs radio wave sensors to measure the power of the radiated energy given off by communication devices and both measures and models how the signal strength varies, within the built environment, due to constructive and destructive interference of the radio waves caused by the multipaths that they travel from transmitter to each point within the environment scattering as they interact with the buildings and other objects to be found there.

The current invention takes that concept further, both in the method of analysis used and by applying it to a much broader range of phenomena that can be sensed, analysed, modelled and recorded.

Figure (1), shows one instantiation of the invention. Feature (1) is the overall housing that might be used to provide a mounting structure and weatherproof shield from both the antennas, electronics and associated power and control circuitry. The instantiation shown as feature (1) is a hemispherical dome upper surface and a cylindrical lower portion of the structure. Other shapes for the housing may also be devised without changing the basic radio sensor equipment or functionality.

Feature (2) is one type of transmitter/receiver device that could perform the required sensing and probing beams that will be needed to sense the range of phenomena in the environment. This type of device is called a Multiple In Multiple Out antenna array, or MIMO for short. It is a very common antenna type for use in the 5G mobile cellular telephone and network systems. It may incorporate the transmitter and receiver circuits and makes use of an array of flat plate antennas either face outwards or edge on to the square or rectangular format. Alternative antenna forms and designs, such as conformal or shaped antennae may replace square or rectangular antennae without changing the functionality and capabilities of the invention. Alternatively, the transmitter and receiver circuits could be housed outside the antenna housing but this is usually less efficient due to energy losses in the radio frequency cables needed to connect the antennas to the radio electronics. Multiple MIMO antenna systems may be mounted in the RF Environmental Sensor facing outwards so as to cover the full hemisphere of this particular instantiation. Other designs that cover the full 4n Steradian of space can equally be envisaged. Feature (3) is an alternative antenna type that may be used in the invention, the one shown in this figure is a high gain Yagi type antenna, although lower gain versions may also be used. Feature (4) is a second alternative antenna design that could be used and it is a helical antenna. Many different types of antenna may be used but the need for multiple antennas to give high gain in many directions is required. This results in the need for arrays of antennas. These antenna array as well as requiring high gain also require to be able to point electronically agile beams for both transmit and receive functions. The figure shows a possible configuration of multiple antennas of the types as in feature (3) and feature (4). The most likely type to be used is the MIMO type although they must also be ultra-wide radio band antennas as well to cover the full range of the radio spectrum as practicable.

Feature (5) is the Radio Tx/Rx transmitter and Receiver circuits, signal conditioning electronics and electronic beamforming system. It also houses the power circuits and RF Environment Sensor control circuits. Feature (6) is the Sensor analysis electronics and main digital computer system that is required to process the digital output from the antenna systems and carries out the analysis of the of the signals so as to determine all of the tasks required for the multiple use cases envisaged in order to represent to manmade and natural environments in which the Sensor is located. A key element of this analysis is the use of Machine Learning techniques which may be used at any stage of the analysis processes employed. The computer is connected to the internet and similar networks in order to provide the end users with the data generated by the Sensor.

Feature (7) shows a stylised manmade generic transmitter of radio frequency power, examples being but not limited to, Cellular Telephone Base Stations, TV and radio broadcasts transmitters, Mobile phone Handsets, Wi-Fi hubs, SSR transponders on aircraft and satellite broadcasts. Feature (8) is a generic representation of the intended electronic receiver and controller of the transmitter and is connected to the internet or similar network, the example given is that of cellular telephone base station. Feature (9) is the internet and/or similar networks used to connect the Sensor to control the Sensor management functions feature (11) and inform the end users feature (10), of the Sensor data.

Figure (2) shows a diagrammatic representation of a built environment such as a city. Feature (1) shows the Sensor in the form shown in Figure (1), a hemisphere on a cylinder within which all the radio receiving and transmitting equipment is housed along with the computer equipment. Only one sensor is shown in Figure (2), however, a plurality of sensors may be deployed across a large complex environment such as a city, port, transportation hub or sports complex or collection of these built environments. Feature (12) is a stylistic representation of a building in the built environment. Feature (13) is a personal device that radiates RF power such as mobile telephone handsets, smart wristwatches and other devices that communicate via radio waves while in use, many such devices can be expected to be in use across the typical built environment at any one time.

Feature (14) is a stylistic representation of a cellular network base station with multiple antenna systems which are in constant transmit mode of varying power levels, a plethora of sizes and locations of base station can be expected. A city or other complex environment will host many such base stations to service the needs of the population and associated service functions. Feature (15) is a stylised representation of the many different devices that use radio data and control links, these devices include the following.

Feature (16) represents a structure in the built environment that is part of the buildings or other structures that scatters radio waves that are incident upon them in ways that are different to the overall building to which it is attached. Examples, include, but are not limited to, roof or wall mounted air-conditioning systems, broadcast satellite receiving antennas and radio transparent surfaces on the building sides. Feature (17) represent radio devices such as switches to turn on lights in rooms when the radio is used to detect the movement of people within the room, garage door opening and closing devices and radar burglar alarms and microwave ovens and a plethora of other such devices not covered in previous items all of who radiate radio waves. Feature (18) represent drones operating in the built environment that radiate radio waves for communication and sensing purposes but also scatter radio waves from other sources and modulate them by means of their rotors and general motion within the Urban Canyon. Feature (19) represents broadcast radio and television channels. Feature (20) represents satellite-based radio sources that transmit signals down to the Earth's surface, these include broadcast TV channels, internet connectivity, navigation signals, radar and commercial and governmental data links. Feature (21) represents aircraft flying over, into and out of airports heliports the built environment that are transmitting radio waves for the purposes of, secondary surveillance radar identification and proximity warning, communications, sensing and other uses. Feature (22) represents a change in the physical environment within the city or complex environment that the end users of the data would require knowledge of, these changes include, but are not limited to following; accidents involving moving vehicles, collapse of buildings, parts of buildings, structures including subterranean slumps that result in surface changes such as holes in the highways and streets, explosions and mass movements of people in the event of accidents and building fires or other motivations. Feature (23) surface transportation systems such as traffic on roads, railways, canals, harbours, seaways, airports and combinations of these in transportation hubs and intersections.

The following 7 figures represent the use cases for the Sensor, some are in passive mode and some are active modes. The latter, modes make use of transmissions by the Sensor. Use of the passive mode is advantageous because it contributes to the detection and tracking of drones free from the cost associated with radio spectrum licences that would be required if transmission were required. By relying on the transmissions of others, the system has a lighter touch on the surroundings and avoids the additional cost of a radio frequency licence.

Figure (3(a)) feature (24) In this and all other use case diagrams shown a stylised sensor is shown as a flat plate MIMO style antenna array that is capable of receiving and transmitting over a wide angle from the face exposed to the use case examples. Other antenna configurations may also be used instead of or in addition to the MIMO antenna. Feature (25) shows a stylised transmitter of radio waves that could have any functionalities mentioned above. Feature (26) is an arrow representing the direct path that radio transmissions would follow from the transmitter to the Sensor antenna of feature (24) this signal will be used to cross-correlate with the received signal via other paths to extract the transmitted signal from any other unwanted signals. This provides system gain to boost the receiver signal above noise and background signals. Feature (27) shows an arrow that represents the radio waves from the transmitter, feature (25), that travels in the direction of the building, feature (12). It is assumed that some of these radio waves impinging of the building will be scattered in many directions some of which will be toward the antenna, feature (24). The radio waves scattered toward feature (24) are represented by an arrow, feature (28). By receiving the two signals, direct and scattered off the building permits the Sensor to detect the presence of the building. If the receiver bean used by the Sensor antenna, feature (24) can be changed from having a wide beamwidth to a narrow one and the radio waves from the transmitter scatter off a significant area of the building then the Sensor may record an image of the building face being illuminated and facing the Sensor antenna. The distance of the building can be calculated if the relative position of the transmitter feature (25) is known to the Sensor and by measuring the difference in time of arrival of the signals that followed the direct and the scattered path from the building. Thus, the building may be sensed and recorded by the Sensor in its mapping of the built environment.

The essential physics of the Sensor in use in these use cases of Figure (3) is the power transmitted by feature (25) and received by feature (24) whichever path or object is involved with scattering the radio waves. This physical process is referred to as the Link Budget and Equation {1} shows the direct path feature (26) from feature (25) to feature (24). P_r = P_t G_t G_r X² S L f (X, R)/ (4n)² R² K T W N {1}

Where;

P_r is the total transmitted power from feature (25)

P_t is the power received by the Sensor's antenna feature (24) via path feature (26)

G_t is the gain of the transmitting antenna

G_r is the gain of the receiving antenna

X is the wavelength of the radio waves transmitted

S is the system gain, including integration gain such as when using active modes, e.g. see Figure(8)

L are the loss factors f(X,R) is a function that represents the attenuation of the radio waves as they pass through the atmosphere and is dependent on the wavelength X and the range R and the atmospheric attenuation along the path between the transmitter and receiver.

R is the range of the Sensor antenna from the transmitter, Feature (26) in Figure (3(a))

K is Boltzmann's constant

T is the absolute temperature of the receiver front end

W is the effective bandwidth

N is the integrated noise factor including the noise in the input stage of the receiver

For Figure(3 (a)) the alternate path from the transmitter feature (25) to the Receiver feature (24) is the situation is often referred to as the 'Bi-static Radar Equation' where the radio waves travel first to the Building feature (12) along path feature (27), scatter off the building feature (12) and some of the wave power follows the path to the receiver feature (28). In this case the power received by the receiver of the Sensor antenna feature (24) is given by the equation {2};

P_r = P_t G_t G_r X² S L o_b f (X, R) / (4n)³ (R₂₇)² (R₂₈)² K T W N {2}

Where the new variables are; o_b is the scattering or radar cross-section of the building feature (12) (for the geometry of the paths feature (27) and feature (28))

R₂₇ is the range from the transmitter feature (25) to the building feature (12) along feature (27)

R₂8 is the range from the building feature (12) to the receiver feature (24) along feature (28)

The Sensor feature (24) then can then determine the range to either the transmitter feature (25) or the building feature (12) if the locations of two out of the three of them are known and the direction of the incident radio waves on the face of the antenna can be measured by the Sensor, a MIMO would be able to do this by forming a narrow high gain (G_r) beam to scan for the signals bouncing off from feature (12).

The remaining use cases in which the Sensor operates in this passive mode, i.e. are similar but involve an intermediate object feature (22) in Figure(3(b)) that scatters the radio wave from feature (12) towards feature (24) for it to detect. The path feature (28) for the radio waves scattering off feature (12) toward feature (24) has been left off Figure (3(b)) for clarity but it can be assumed that it is present and the combined illuminations of feature (12) and feature (22) both take place and the Sensor can detect and determine the location of both. Although there will be ambiguity about the location of feature (22) unless feature (24) can point a receiver beam at feature (22) as well as feature (12) and resolve them both simultaneously. The difference in arrival times of the radio waves from feature (25), feature (12) and feature (22) at feature (24) are used in the determination of the locations of them relative to feature (24).

Figure (3(b)) shows a similar situation to that of Figure (3(a)) but with the addition of the feature (22) the change in the environment that now scatters some of the radio waves being scattered off the building, feature (28) represents some of the radio energy from feature (25) scattered of feature (12) that is now falls upon feature (22) and is scattered toward the antenna of the Sensor as shown by the arrow of feature (28). In this instance the Sensor can form an image of the building and the object or change represented by feature (22) but this time the distance and position of the object can be measured and recorded by the Sensor as well as the building. This being achieved even if there was no direct illumination path between the transmitter and the object.

The remaining use cases in which the Sensor operates in this passive mode, i.e. are similar but involve an intermediate object22) in Figure(3(b)) that scatters the radio wave from feature (12) towards feature (24) for it to detect. The path feature (28) for the radio waves scattering off feature (12) toward feature (24) has been left off Figure (3(b)) for clarity but it can be assumed that it is present and the combined illuminations of feature (12) and feature (22) both take place and the Sensor can detect and determine the location of both. Although there will be ambiguity about the location of feature (22) unless feature (24) can point a receiver beam at feature (22) as well as feature (12) and resolve them both simultaneously. The difference in arrival times of the radio waves from features (25), (12) and (22) at feature (24) are used in the determination of the locations of them relative to feature (24).

Figure (3(c)) shows a similar situation as in Figure (3(a)) but in this case the scattering object is not a stationary building but moving surface traffic, feature (23). The scattered radio waves will give the position and overall shape of the traffic flow but this time the radio waves are scatted by moving traffic. This causes the scattered radio wave to change their frequency as in the Doppler effect. This effect can be used in a number of ways by the Sensor. Firstly, it can be used to calculate the speed of the traffic in the radial direction away from or towards the Sensor. This may also be used to track the individual vehicles along the road if the range and direction of the road is available in the Sensor memory. Finally, the Sensor may generate the equivalent of a Waterfall Plot within its computations. This allows the length of the vehicle to be measured since the degree of Doppler effect experienced by the target vehicles illuminated by the transmitter depended on the geometry of the illumination and sensor directions relative to each other. Thus, for a long vehicle such as a bus the Doppler frequency shift for the front will potentially be measurably different to that at the rear of the bus. This is useful information in situations where the source of the transmissions is at longer wavelengths such that narrow beams cannot be formed by the Sensor to image the target vehicle. Thus, the Doppler spread caused by the target length can distinguish a motorbike, from a car and a car from a bus. Other transport systems and devices can be imaged and/or categorised such as trains, ships and aircraft by these simple measures.

In figure (4) the situation is similar to the previous use cases in that the MIMO antenna, feature (24) detects the radiated radio waves feature (26) from transmitter feature (25). This time however the radio waves feature (27) from feature (25) that are considered are incident on the surface of the terrain. In this case a body of water, feature (30), has overflowed and the excess water, feature (31), is flooding across the terrain and causing material damage to both the natural environment and the human built environment. The radio waves feature (27) are incident on debris in the form of a partially floating building and tree feature (32) and also waves and ripples in the surface of the water. The scattered radio waves feature (29) from these objects and the general surface water could be Doppler shifted and thus contain useful information about the flow rate and direction that the Sensor may detect and analyse. The water surface, even if flat and not carrying debris, could be detectable in comparison to the underlying terrain which may have been previous mapped by the Sensor as part of a baseline model of the terrain.

Figure (5) shows a similar situation to that of Figure (4) but in this case the natural phenomena causing the reflected radio waves is the appearance of a Tornado, funnel cloud or water spout feature (33) and the debris it carries. The Doppler shift produced by the scattering will be different for different parts of the Tornado as the rotating cloud and its debris will have components of speed ranging from travelling directly toward the Sensor feature (24) and directly away and all velocities in between. The velocity of the overall Tornado as it moves across the terrain will also produce a component of velocity toward or away from the Sensor and this too will result in Doppler shifted signals that the Sensor may be able to detect and record.

Figure (6) shows a similar arrangement of transmitter feature (25) and the Sensor feature (24) and the radio waves, features (26), (27) and (29). This use case, however, involves the detection of the effects during and after an Earthquake. Fissures in the ground and changes in topological height such as represented by feature (34) could be detected by the Sensor feature (24) both by any Doppler shift in the scattered radio waves from that segment of terrain and by comparison to the previously sensed and modelled baseline data. The effects of Earthquakes on manmade objects such as a lamp post feature (35) as it vibrates and rocks due to the shock waves of quake will be detected by the Sensor as Doppler shifts in the frequency of the radio waves scattered off the objects and also if any structures or buildings collapse or are moved in position relative to their pre-quake locations. The Sensor will take into account the possible shaking of both itself and the transmitter used in these detections as well as any translational movement of either. This being achieved by both inbuilt vibration sensors in the Sensors and by refence to transmitters that are not subject to the Earthquake effects such as those mounted on overflying aircraft that may be present at the time or more distant transmitter such as Satellites and ground-based broadcast TV and radio transmitters distant from the Earthquake zone. The Sensor will process such data collected during and after the Earthquake and record it into the baseline data.

Figure (7) shows Sensor and Transmitter arrangement similar to the previous figure but this time being used in the use-case where weather phenomena in the form of rain clouds feature (36) and lightning strikes feature (37). In the case of the clouds, the radio waves would be scattered by both the water droplets in the cloud and the rain drops falling from the base of the clouds. The Doppler shifts caused by the motion of the clod and the falling rain drops will be used to detect and determine the rate of fall and used to indicate dangerous downpours and Microbursts air movements and turbulence. Warnings to aircraft, airport and air traffic control authorities would be released by the Sensor control system feature (11) in Figure (1).

Figure (8) shows four use-cases that are similar in the geometries to previous use cases but differ significantly now in that the Sensor is operating in 'Active' modes where the Sensor itself transmit radio waves as part of its functionality. In Figure (8 (a)) the Sensor feature (24) and Transmitter/Receiver unit feature (25) which can be but not limited to a Cellular Telephone Base Station, are both able to communicate with each other using transmission and receive channels of radio waves. The Sensor feature (24) first emits radio waves feature (38) directed towards building feature (12). These radio waves are scattered by the building feature (12) and some of the radio signals feature (39) then falls on the Transmitter/Receiver feature (25). The transmissions from the Sensor feature (24) could for example be of a nature normally sent from Mobile Telephone handsets, or other such devices normally found in modern human use. In this use case the Transmitter/Receiver feature (25) is a taken to be a Cellular Telephone Network base station and processes the signals feature (39) as a legitimate call that it sends on into the network to an address contained within the call feature (38) sent by the Sensor feature (24). This destination address will be part of the Senor's electronic system and so the Sensor will be communicating with itself via base station feature (25) and the associated Cellular Network that it serves. The purpose of this communication with the Sensor by itself is to send signals that can be used to obtain more details and at higher resolution that just relying on transmitters of opportunity that are available when the Sensor needs to obtain important data on the environment. The use of narrow transmit and receive beams by both the Sensor feature (24) and Transmitter/Receiver feature (25) as in 5G Cellular Networks will focus the power on the building and maintain the safe standards required by law. However, the fact that the Sensor now has a communication loop that it can send modulated signals means that extra processing gain can be obtained and allow far greater ranges for detection of the returned signals at the Sensor and thus obtain greater detail over a larger area of the environment surrounding the Sensor. Further details of this enhanced performance will be discussed later when considering Figure (11).

Figure (8(b)) considers the use-case where an object or small change to the environment needs to be examined by the Sensor feature (24). In this case use is made of a communication link in exactly the same way that the communication link by the Sensor with itself via Base Station feature (25) was created in the use case considered in Figure(8 (a)). However, the radio waves from both the transmitted signals from features (24) and (25) fall on (12) and scatter off feature (42) the object/change feature (22) and are received by the Sensor by Radio waves feature (43). Once again, the modulation of the communications from and to feature (24) enable greater performance for the Sensor in its detection, location and possibly imaging of feature (22).

Figure (8(c)) shows an identical situation to that of Figure (8(b)) minus the base station feature (25) but where the object in question is now a moving stream of traffic illuminated by bouncing a transmission off the building feature (12). The positions and motions of the traffic may be determined as in previous use-cases but this time the illumination is purely provided by the Sensors and can also make use of the enhanced performance the Sensor due to the use of signals of known modulation. Once again imaging of the traffic flow will be possible by use of fine beam widths and scanning of the MIMO antenna option for feature (24) particularly at the higher frequencies and shorter wavelengths available in 5G and future standards.

Figure (8(d)) shows the use case where the object is now a building and this represents the ability of the Sensor to scan, image and locate in 3D the entire city or complex environment surrounding it using the techniques described in previous use cases above.

Figure (9) shows three new use cases not covered by the previous ones above. Figure (9(a)) shows the case where an aircraft feature (21) is transmitting signals feature (26) to the Sensor and feature (27) toward the terrain below it. These signals need not be specifically to aid the Sensor feature (24) but could be general emission typically made by modern aircraft such as SSR signals to identify itself to air traffic control radar systems, other such emissions may also be used by the Sensor. The specific mode of operation of the Sensor that is represented is that of Bistatic Synthetic Aperture Imaging. The transmissions from the aircraft feature (27) are received by the Sensor feature (24) directly. Also, the transmitted signals feature (27) from the aircraft feature (21) scatter off the terrain feature (28) toward the Sensor. The sensor can then process these two sets of signals taking into account the differences in times of arrival between the direct signals and the signals scattered as well as the directions from which both sets of signals are arriving. Using standard Bistatic Synthetic Aperture Imaging algorithms a map of the terrain below the aircraft may be computed by the Sensor. In the case of SSR signals the altitude, speed and direction of the aircraft is encoded into the transmissions and the Sensor can decode these to retrieve the information. Cross correlation between the direct signal feature (26) and the scattered signal feature (27) will be used in the mapping algorithms.

Figure (9(b)) shows the situation where a legitimate source of signals is being used, in this case from a satellite navigation system in orbit feature (20). These radio waves from feature (20) may be received by the Sensor feature (24) directly feature (26). If an illegal transmitter feature (49) is also transmitting on the same frequencies as the navigation system feature (20) in an attempt to jam or spoof the navigation system signals then the Sensor feature (24) may be able to receive these illegal signals by both a direct path, which the MIMO antenna will be able to determine the direction of feature (49) but not its range and therefore will not be able locate the jammer/spoofer feature (49). However, if the signals from feature (49) are also falling upon a building feature (12) which is in a known location and the jamming/spoofing signals feature (51) also fall upon the building feature (12) and are scattered off as in feature (52) in the direction of the Sensors then the Sensor will be able to locate the position of the jammer/spoofing transmitter feature (49). If more than one building is scattering the illegal signals then the accuracy of the location may increase. The sensor will then be able to notify the authorities via its Command centre feature (10) in Figure (1) and appropriate action take to stop the illegal transmissions. Other illegal sources of radio waves may also be detected and located by this method.

Figure (9(c)) shows the situation where the Sensor is monitoring the power of radio waves emitted from multiple sources feature (53). These can be across all bands and from all emitter types and provide a diagnostic tool to measure the 'Health' of the radio community. In the event that signal traffic changes then it may indicate that unusual events are taking place and the authorities and the emergency services may be alerted to the changes from normal. Transmission power levels dropping out could indicate power supplies to areas have been lost or if unexpected greatly increased traffic on Cellular Networks for example is taking place then some form of emergency may be in progress. Other factors, such as changes in the intensity of the Solar Wind, or Solar Coronal Mass Ejections which can greatly affect many radio wave multiple sources feature (53) could then be detected early and measures implemented via the Command centre feature (10) to protect systems and vital devices such as the Critical National Infrastructure.

The radio sources that are proposed to be used as RF sources and illuminators include, but are not limited to the following.

Radio and TV Broadcasting: - Radio bands VHF and HUHF, powers transmitted range from, 100,000s of Kilowatts to several Megawatts. Signal formats are Analogue and Digital.

Mobile Phone Base Stations: - The networks employed as Cellular Telephone communication systems for both public and private networks employ radio transmitters and receivers and combined transmitter receiver units in their architectures. These sources of radio waves and their receiver functions are use singly or a plurality of them by the this invention to sense the physical environment in the areas around them. These networks may include individual home, office or other venues that use WiFi networks to connect devices within the venue to the internet. The radio transmissions from these WiFi devices may be used singularly or a plurality of them by the System to detect both changes in the use of the networks such as overall demand for calls to be established as well as the physical environment around them.

Mobile Phone Handsets: - Mobile phone handsets that are used in cellular telephone communication networks may be used in this invention in either cooperative mode where the handset transmits and or receives RF signals for the purpose of facilitating this invention's ability to sense the environment. Alternatively, the transmissions from the Mobile phone handsets that come from their normal non-cooperative use may be used by the System to judge the demand for communication in the locations from which the handsets are operating as well as in sensing the environment around them. Included in this can be the cellular telephone units that are not hand held but are incorporated in items regarded as part of the 'Internet of Things' (loT). Such loT transmitters and receivers operating in cellular networks may be used by the System in both cooperative mode as well as non-cooperatively as with handsets. Drones also make use of the cellular telephone technologies and networks for control and the transmission of image and other sensor data to their control and other functions on the ground or elsewhere and the Sensor may make use of these also.

RF transmission from Aircraft: - Commercial and military aircraft usually transmit RF signals for a variety of reasons and the Sensor may make use of these transmissions. The use of Secondary Surveillance Radar (SSR) transponders in aircraft of all types is very common and of particular use to the System for sensing the environment around it. These transmissions are encoded with information about the aircraft identification, position, flight level and other data that may useful in the sensing of the environment calculations when the Sensor uses them and their scattered, time delayed and Doppler shifted multipath signals. The Sensor may use the SSR RF transmission to form images of the environment along the flightpath of the aircraft by use of the technique known as Synthetic Aperture Radar and also Inverse Synthetic Aperture Radar.

Space based RF Transmissions:- RF transmissions from manmade satellites in Earth orbit and beyond may be use singularly and in plurality by the Sensor to sense the environment in SAR, ISAR and multipath imaging in a similar way to the other RF sources described already in this patent. Natural space-based RF transmissions may also be used by the Sensor to build up over time maps of the environment by virtue of the techniques described herein and integrated by many repeated transits of the astronomical sources such as the Sun, Jupiter and Pulsars making use of their known positions on the celestial sphere. The remaining 7 figures describe the functional architecture of the Sensor.

Figure (10) shows the top-level functionality of the Sensor.

Feature (54) System Management: This function would permit both human invention and automatic computerised control of the Sensors.

Feature (55) RF Sensor 1: The Radio Frequency (RF) used are made up of a plurality of radio antenna arrays and a plurality of electronic modules that may transmit and receive the energy of the radio waves via the antenna arrays, details of these RF sensors are provided in Figure (11).

Feature (56) RF Sensor 2: is the same nature as RF Sensor 1.

Feature (57) RF Sensor of N sensors: Where N is a plurality of RF sensors and are the same nature as RF Sensor 1.

Feature (58) Internet Sensors (TV, IR, Acoustic+): These sensors are accessed via communication networks such as, but not limited to, the internet and may be integral and exclusively for the use of the Sensor or provided for other purposes by independent parties and may be a plurality of sensors of a plurality of types and purposes.

Feature (59) Sensor Processing: The outputs of the RF Sensors are connected to the Sensor Processing unit which is described in detail in Figures (12) and (13).

Feature (60) Baseline Environment Model: The baseline Environment model is a comprehensive data set held in computer memories both within the Sensor and elsewhere, whether for data availability reasons, distributed processing reasons or for backup purposes, it is used to compare the output from the live Sensor output to the Baselines content and generate a report that categorises the differences between the two data sets, its functionality is described in Figure (15).

Feature (61) Baseline Environment Model Update: this functionality is the output of the Sensing process and is used to compare with the Baseline Model.

Feature (62) Anomaly Morphological Evolution: the output of the Sensor Processing function contains data on the size, shape, location, motion and changes to those over the observation period measured and detected and reported in the Sensor processing output.

Feature (63) Anomaly Detection Report: this report is generated when the Sensor Processing function produces its reports for individual sensors and is a correlation of reports from the individual sensors reporting measured changes to the environment they have detected. Feature (64) Anomaly Motion Tracking: this functionality combines current and previous detections of anomalies and forms tracks of anomalies that appear from the data to be related to the same anomaly but in a different location in a way similar to radar tracks being generated from a history of radar plots.

Feature (65) Anomaly Categorisation: This functionality examines the data on correlated and individual non-correlated anomalies using a variety of standard techniques including pattern matching and more sophisticated Machine Learnt Algorithms that are of proprietary nature.

Feature (66) Preliminary Environment Model Update: A preliminary set of changes to the Environment Model are collected together in the form of a report.

Feature (67) Action Decision: The decisions concerning the inclusion of Environment Model changes and including them in the Update to the model are handled both by automatic computer based algorithms and where the changes are deemed significant by predetermined inbuilt criteria referred for human decision making.

Feature (68) Interface to End Users: Where significant changes have been detected in the environment these will be sent as messages to the group of human and automatic algorithms that constitute the end users of the Sensor output, these include but are not limited to; local and national authorities, transportation system management, emergency management agencies and first responders, communication network management, power distribution management and the public.

Feature (69) Environment Model Management: this functionality is routine management of the Environment Model and includes regular backup of the model as well as the updates to it.

Figure (11) shows the functionality of the individual RF sensors.

Feature (70) RF Antenna Array 1 and 2; each RF antenna array, see Figure (1): features (2), (3) and (4), may be utilised individually or in unison by feature (73) and this includes both in Active, i.e. transmission mode, Passive, i.e. receiver mode, or in transmit and receive modes.

Feature (71) RF Antenna Array Mth: is of the same nature as feature (70).

Feature (72) RF Sensor Management and Tasking: here the functionality includes both routine sequences of utilisation of the RF Sensors antenna arrays, RF modules and antenna switches and non-routine utilisation when significant anomalies have been detected in the environment, in the latter case these may be automatic computer controlled or human controlled interventions. Feature (73) Radio Modules and Antenna Switches: These functionalities are carried out by the hardware, firmware and software within feature (5) in Figure(l) and carry out the transmit and receive functions switching the signals to and from the antenna arrays.

Feature (74) Receiver Control: This functionality carries out the instructions from feature (72) and controls the RF receiver functions

Feature (75) Transmitter Control: This functionality carries out the instructions from feature (72) and controls the RF transmitter functions

Feature (76) Beam Forming Control: This functionality carries out the instructions from feature (72) and controls the formation of the beam shapes and directions using, for example but not limited to, digital beam forming algorithms.

Feature (77) Direct or Transmitted Path Signal: this functionality takes the time sampled RF signals that have been received from external transmitters in the environment external to the Sensor and then conditions them by adjusting the amplitude and other processes such as but not limited to filtering the signals in the frequency domain.

Feature (78) Scattered Path Signal: this functionality takes the time sampled RF signals that have been received from both the Sensors own transmitted signal and/or external transmitters that have been scattered by objects in the environment external to the Sensor and then conditions them by adjusting the amplitude and other processes such as but not limited to filtering the signals in the frequency domain.

Feature (79) I and Q Outputs: This functionality takes the RF signals from features (77) and (78) and separates them by means of one of a set of standard processes to create the In-phase and Quadrature, i.e. I and Q, formats for each of the RF signals and conditions them in amplitude before outputting them.

Feature (80) Signal Cross correlation: This functionality takes the two sets of signals samples from features (77) and (78) and converts them to digital signals then cross-correlates them together to determine the time delay between them which is used to determine the range of the object that scattered RF signals. It output both I and and time delay data.

Figure (12) shows the Sensor processing function, feature (59) of Figure (10).

Feature (81) RF I and Q Signal Input: this function distributes the RF signals to the Advanced Signal Processing functions. Feature (82) Internet Sensors (TV, IR, Acoustic+): this function distributes the images and signals to the functions.

Feature (83) Advanced Signal Processing: This takes I and Q. signals and is shown in detail in Figure (13).

Feature (84) Scattering Centre, Image Correlation and Anomaly Detection: This takes the output of the Advanced signal processing function, RF I and Q. signals and the internet sensors data and detects anomalies when compared to the Baseline Environment Model and is shown in detail in Figure(14).

Feature (85) Distribution of Data to Anomaly processing functions: This function sends the relevant data to each of the anomaly processing functions in features (62), (63) and (64) of Figure(lO).

Figure (13) shows the Advanced Signal Processing.

Feature (86) I and Q. signals and time delay data digitisation and distribution: in this function the I and Q. signals are sampled and digitised by A to D converters and then distributed to the processing functions.

Feature (87) FFT/DFT Frequency Domain and Phase Calculation: the processing produces power spectra in the frequency domain and corresponding phase/frequency curves, advanced algorithms then analysis these plots to identify significant features, such as, but not limited to, particular types of RF sources and their associated signal powers, wave bands and data formats.

Feature (88) Wavelet Analysis Calculation: the processing produces information similar to FFTs and DFTs but can usefully analyse non-stationary (i.e. changing in time) signals and extract information on the time dependent characteristics of the signal.

Feature (89) Cepstrum Analysis Calculation: is useful in detecting moving objects in bistatic RF location systems and would be used in, but not limited to, road and rail traffic flow analysis.

Feature (90) SAI and ISAI Imaging: Synthetic Aperture Imaging involves a moving RF source providing both direct path signals to the Sensor and also indirect path signals that have been scattered off of the terrain and manmade features on it, whereas ISAI is Inverse Synthetic Aperture Imaging and allows an image to be formed when a radio waves have been scattered off of a moving and/or rotating object from an RF source that is of known location, both imaging functions are well known and the algorithms used are open source information. Feature (91) RF Tomographic Imaging: The tomography algorithms commonly known from its medical applications are adapted in this function to operate on radio waves passing through objects such as but not limited to buildings and forests to reveal the internal structures and scattering centres.

Feature (92) Collection of Signal Information: in this functionality the outputs of the advanced signal processing functions are correlated for distribution and use by the feature (84) of Figure (12).

Figure (14) Scattering Centre, Image Correlation and Anomaly Detection. This processing function collects signal, images and other sensor data and processes and analyses the data in order to detect anomalies in comparison to the Baseline Environment Model.

Feature (93) Input from RF Signal information: this function takes data from the RF stream of signals and connects it to the signal conditioning function feature (97).

Feature (94) Input from Imaging (TV and IR), Acoustic+ Sensor(s): this function takes data from the Internet and other networks providing TV, IR, Acoustic plus other types of sensors as are available and image correction function feature (98).

Feature (95) Input from Advanced Signal Processing: This function provides data from the Advanced signal Processing function feature (83) of Figure (12).

Feature (96) Input from Baseline Environment Model: This is the function feature (60) in Figure (10).

Feature (97) RF Signal Corrections: this function corrects the amplitude and time stamping of the RF data.

Feature (98) TV and IR Image Corrections: this function applies corrections to, but not limited to, contrast, brightness and colour.

Feature (99) Baseline RF Component Environment Model: this function extracts the RF generated components from the Baseline Environment Model.

Feature (100) Baseline TV and IR Component Environment Model: this function extracts the non-RF generated components from the Baseline Environment Model.

Feature (101) RF Scattering Centre Calculations: This function converts RF digitised signals into 3D locations of scattering centres.

Feature (102) TV and IR Image Analysis using Machine Learning: This function uses advanced Machine Learnt feature extraction and classification algorithms. Feature (103) Integration of Scattering Centres and Images Features: this function collects and correlates the data on features detected and located in all sensor data types and correlates them into anomalies for further analysis.

Feature (84) Scattering Centre, Image Correlation and Environment Anomaly Detection: - This function correlates data features and defines them as anomalies or non-anomalies for further analysis of the anomalies and is also shown in Figure (12)

Figure (15) this function where the Baseline Environment Model, feature (60) of Figure (10), is generated and maintained.

Feature (105) Input from Environment Model Update: this function is feature (61) of Figure (10).

Feature (106) Time Sequenced 3D CAD Model of Complex Environment: This is the multidimensional computer model of all phenomena modelled and measured of the environment surrounding the Sensor and is organised into time sequenced slices that cover the cycles of human and natural activities recorded and modelled up to the instant of the particular 'Present time'.

Feature (107) Static RF Domain Model Generation: This data set has been generated by computer modelling of the static features of the environment and includes all known sources of radio transmissions and the interactions with the static features, such as, but not limited to, buildings, bridges, harbours, airports, tunnels entrances, trees, lakes, rivers, coastlines and topography, it defines the frequencies, power levels, polarisations, signal formats and directions of flow against the time in the cycles of the Patterns of Life of the human inhabitants.

Feature (108) Dynamic RF Domain Model Generation: this function is the data set for dynamically changing features in the environment such as, but not limited to, road, rail and drone traffic, it includes the normal location of the flow in 3 dimensions, its velocity range, radio frequency crosssection distribution and signal strength for each static RF source.

Feature (109) Visible and IR Light Domain Model Generation: This data set includes known manmade light and thermal sources in the environment as well as natural light and heat under a typical range of weather conditions over the diurnal and annual and lunar cycles illumination of the environment and its scattering within it.

Feature (110) Infra-Red Domain Model Generation: This data set includes all known manmade IR sources, such as associated with security cameras as they are distributed over the environment the illumination they provide in the objects and terrain of that environment. Feature (111) Acoustic Domain Model Generation: this data set includes all known manmade and natural sources of sound in the environment and the reflection attenuation and absorption by the environment expressed in sound pressure level, frequency over the pattern of life cycles.

Feature (112) Geophysics Domain Model Generation: This data set includes all known geological features such as water courses, fault lines, areas subject to subsidence, volcanoes and geysers and their natural cycles.

Feature (113) Meteorological Domain Model Generation: this data set includes the known ranges of weather patterns in the location of the Sensor of the natural cycles as well as predictions from modelling of the wind effects within the environment as measured and predicted according live feeds of metrological data and weather forecasts.

Feature (114) Biosphere Domain Model Generation: This data set includes plant growth, including agriculture, wild animal, birds, fish and insect patterns of movement.

Feature (115) Hydrological Domain Model Generation: data set on the natural flows of water in the environment including overflows and flooding patterns including the extremes of the natural cycles.

Feature (116) Synthesis of Baseline Environment Model: this function collects the data for all categories that can be expected at the time of day, day of the week and year under normal circumstances.

Feature (117) Interface to Sensor Processing; this function transfer current Baseline Environment Model throughout the Sensor functional elements.

When constructing the baseline model of an environment, the location and dimensions of key items such as buildings, lamp posts and other street furniture are taken into account. The baseline model therefore identifies parts of the environment which are never capable of reaching the predetermined criteria for safe passage of a remotely operated vehicle.

For most environments, the locations of these items are usually taken to be fixed, but in areas of elevated tectonic plate activity and areas subject to extreme weather events, even the location of buildings and land topology is monitored. This is especially important if the baseline model is made available to a search and rescue drone seeking to locate and assist in the aftermath of a landslip, flood or other natural disaster. The remotely operated vehicle cannot assume that the topology of the land remains unchanged and therefore it must be monitored in addition to other aerodynamic and electromagnetic factors that are discussed in more detail below. Beyond the parts of an environment that would bring a remotely controlled vehicle into contact with a static object or piece of topology, i.e. a building or a hill, there are other no-go areas within an environment which are dictated by aerodynamic conditions or electromagnetic conditions or a combination of these, and other, conditions. These conditions are constantly changing and are subject to monitoring.

Aerodynamic factors

A remotely operated vehicle cannot enter an area that is too close to a building. In the close vicinity of a building an aerodynamic boundary layer is created that can result in very high local wind speeds. Street furniture such as lamp posts can also create exceptionally strong turbulent eddies which dynamically shed.

In the vertical plane, the lower limit of the environment in which the criteria for safe passage can ever be met, must be above the level of the tallest vehicles moving along the ground at the time. Furthermore, close to the ground there is more turbulence and vorticity which can create swirling effects.

In the vertical plane, the upper limit of the environment in which the criteria for safe passage can ever be met, is typically aligned with the tops of the buildings in the urban environment. The further above the ground the higher the wind speed, in general. As a result, buoyancy and sheer layers can cause highly challenging aerodynamic conditions.

Electromagnetic factors

Some parts of the environment are in line of sight of one or more electromagnetic transmitters. These transmitters may be mobile phone masts. In some parts of the environment, there is no line of sight to a transmitter, but there is sufficient reflection from buildings to provide a reflected signal that indirectly impinges on the environment via multipath.

Some environments or parts of environments may have destructive interference which results in a lack of signal. These areas are sometimes called "not-spots" as the converse of the situation where a "hot-spot" is created as a result of constructive interference. Some physical features of the environment, such as lamp posts, are metal. Metal objects can distort the electromagnetic signal in their vicinity. The exact nature of the modification of the electromagnetic signal will depend on the frequency and wavelength of the electromagnetic signal and also on the size and orientation of the street furniture relative to the incoming electromagnetic signal.

Although the buildings within the environment will remain in the same place over time, at least in the majority of situations, the properties of the building will change depending on the weather. The presence of water on a surface will considerably alter the reflectivity of a surface and therefore the presence or absence of water from the surface will contribute to measurable changes in the electromagnetic signal in the environment. For example, dry tarmac absorbs radio frequency electromagnetic signals whereas wet roads and cars, whether wet or dry, reflect these signals and create multi path interference.

The monitoring of electromagnetic factors may be entirely passive and focussed on the sensing of the electromagnetic signal landscape and identifying the location of hot spots and not spots and providing this information as one of many dynamic features capable of modifying the baseline model.

Pattern of Life

This provides a context to the baseline model as some factors that influence the environment may have predictable cyclic behaviour on a daily, weekly or yearly basis. For example, more vehicles will be expected on a road at morning and evening rush hours Monday through Friday, whereas fewer vehicles will be expected on a Saturday or Sunday. Key local holidays such as Christmas, Eid and Thanksgiving will be included in the baseline model too because they create different behaviours.

Data Fusion

The baseline model is created as a fusion of all of the available factors including the pattern of life data; aerodynamic sensed data; electromagnetic sensed data and geometric or topological data.

When identifying whether or not it is safe for a remotely operated vehicle to enter an environment, certain criteria must be met, i.e. certain predetermined threshold levels of these criteria must be exceeded. For example, the radio frequency signal strength, which is required for control signals to pass to the remotely controlled vehicle, must exceed a predetermined threshold throughout the environment. In other words, the intended path of the vehicle through the environment must not encompass any not-spots of radio frequency signal.

Some drones rely on continuous or near continuous data transfer from a remote location to the drone itself and therefore one of the criteria that must be met is that there must be continuous connectivity for the drone to its signal provider. This is a more stringent requirement than for a voice call on a mobile for example, in which the data is packaged and different packets may take different routes and as long as the majority of packets arrive there is sufficient to sustain a meaningful connection. For a drone, the loss of one or more packets could be significant and therefore the predetermined criteria required for safe passage through an environment includes an acceptable level of packet loss which will be low or may be zero.

Furthermore, in order to achieve continuity on a voice call when a user is moving, it is very common to switch the base station from which the user receives their signal. However, the switching time can be unacceptably long for a drone which may be moving at speed through the environment and cannot accept loss of signal for the time that would be taken for the signal provider to be switched to another base station. One of the criteria that must be met to enable safe passage can therefore be to ensure that hand offs between signal providers are minimised and, where needed, can be carried out sufficiently quickly that signal is not lost in the transfer.

Active steering of electromagnetic factors

In the instance in which the relevant electromagnetic frequency range is that of mobile phone, then active transmissions can be made by a mobile phone in order to stimulate the transmitter to direct its electromagnetic transmissions in response to the mobile phone request, in the direction of the mobile phone. The transmitters are steerable and are typically steered to ensure that the majority of user requests can be met at any time. By raising a request from an environment, the electromagnetic factors in that environment can be changed. Once in communication with the transmitter two way digital coded modulation can be used to cross-correlate with the locally generated version of the code as set out above with reference to feature 24 of figure 3A. This steering can be important in boosting the signal into an urban environment to ensure that there is sufficient signal, through multipath reflections, down through the depth of the urban canyon. These augmented electromagnetic factors are fed into the baseline model through data fusion and can boost the model to meet or exceed the predetermined criteria for safe passage of the remotely operated vehicle through the environment.

The steerability of these transmitters can negatively impact the baseline model in circumstances where a considerable number of requests from a different environment cause the transmitter to steer away from the intended path of the remotely operated vehicle. For example, when a train carrying a considerable number of users arrives in an environment, all of the users leaving the train can start to stimulate the transmitter to direct the electromagnetic transmissions towards them. In doing so, and meeting the needs of the majority, the electromagnetic signal strength in environments non-adjacent to the station at which the train has arrived, will be down selected for coverage as the transmitter steers to serve the majority request.

The environment may be of any suitable size. For example, an environment may be a whole city scape. Within an environment, there may be many smaller environments, each of which may be identified as safe for a vehicle or not safe at any given time. Each environment may overlap with one or more adjacent environments in order to provide continuous coverage of a vehicle's intended journey.

Figures 16A to 16E show five separate traces showing the dynamic fluctuations of the electromagnetic signal over time. These are data obtained in a complex urban environment that does not have direct line of sight from a transmitter. As a result, the signal present in the test environment is formed from a combination of diffraction over the edge of adjacent buildings and reflection from a large building opposite the sensor.

Each plot is divided into an upper and lower plot. The upper plot shows the signal intensity across a 3G frequency band from 2.150000 to 2.155000 GHz. A 3G base station was chosen because it has a wide bandwidth in which to observe interference. The thin line in the upper part of the plot shows the maximum power from the base station since the commencement of observation. The broad band in the upper plot shows the integration of the signal for a given time period. This information is then converted into the waterfall plot shown in the lower image. The figure provides a snap shot of this data which "flows" downwards across the lower plot over time. The intensity of the signal is identified by the shade. The vertical line shows the sampling in phase and in quadrature. The signal strength can be obtained from the root of the sum of the squares.

In Figure 16A, a direct set of data from a transmitting base station is shown. There is a largely homogenous response across the lower plot. The darker striations are believed to arise from constructive interference from reflections from cars and other road traffic passing through the environment. In the upper plot the broad band of integrated response is close to the maximum line showing that most of the available signal is received.

In Figure 16B, the sensor is pointing into comparatively un-complex space with the building edge in the right half of the beam. The back scatter from the edge results in four distinct oscillations across the frequency band. The lower plot also shows these four oscillations. It is interesting to note that the darker, not-spots, do not remain at a constant frequency but move slightly over time. This demonstrates the difficulty of avoiding the not-spots as their frequency does not remain constant. It is also notable that the board band of integrated signal in the upper plot is considerably lower in intensity than the maximum line.

In Figure 16C, the sensor is pointing towards a large building that has a front face than is tipped back at about 5° from the vertical. The centre of the beam is pointed at the building about 1.3 of the way from the edge of the building. Furthermore, this field of view also includes parked cars, which reflect the electromagnetic signals and create considerable multipath. Reviewing the upper and lower plots of Figure 16C, the complexity of the environment is mirrored into the complexity of the plot. There is a hot spot in the lower part of the frequency band with an intermittent adjacent not- spot. The right side of the plot is extremely complex with not spots moving through across the frequencies in a non-uniform manner.

In Figure 16D, the sensor is pointing predominantly at the large building, although not directly facing the building. This plot is more similar to Figure 16B, but with slightly more complex movement of the not-spots over time.

In Figure 16E, the sensor is pointing to the right hand edge of the building which is further away than the left hand edge of the building as shown in Figures 16B and 16C. Beyond the large building there is comparatively simple space where there is nothing to reflect the beam back. As a result there is a reduced range of distances travelled by the signal and the response is more consistent both across the frequency band and in time on the waterfall plot. The signal strength is also higher than in Figures 16B, C and D.

Various further aspects and embodiments of the present invention will be apparent to those skilled in the art in view of the present disclosure, "and/or" where used herein is to be taken as specific disclosure of each of the two specified features or components with or without the other. For example, "A and/or B" is to be taken as specific disclosure of each of (i) A, (ii) B and (iii) A and B, just as if each is set out individually herein. Unless context dictates otherwise, the descriptions and definitions of the features set out above are not limited to any particular aspect or embodiment of the invention and apply equally to all aspects and embodiments that are described. It will further be appreciated by those skilled in the art that although the invention has been described by way of example with reference to several embodiments, it is not limited to the disclosed embodiments and that alternative embodiments could be constructed without departing from the scope of the invention as defined in the appended claims.

Claims

1. A method for improving the safety of remotely controlled vehicles in a dynamic environment, the method comprising: accessing a baseline model of an environment; monitoring at least one dynamic feature within the environment; modifying the baseline model to incorporate the at least one dynamic feature being monitored; determining whether the modified model meets a predetermined set of criteria; and sending an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

2. The method according to claim 1, further comprising the step of generating a baseline model of an environment.

3. The method according to claim 1 or claim 2, wherein monitoring at least one dynamic feature within the environment comprising monitoring at least one of: airflow; water courses; electromagnetic signal strength; light conditions; precipitation; and the location of moveable objects.

4. The method according to claim 3, wherein the dynamic feature is the electromagnetic signal strength and wherein the monitoring of the feature includes identifying the location of the transmitter of the electromagnetic signal.

5. The method according to claim 4, wherein the output further comprises a notification to relevant authority that the monitored electromagnetic signal is indicative of a jamming/spoofing transmission from an identified location. 34

6. The method according to any one of claims 1 to 5, wherein the baseline model comprises the location of at least one static feature.

7. The method according to any preceding claim, wherein the baseline model is modified in real-time or near real-time.

8. The method according to any preceding claim, wherein the output is configured to: allow the vehicle to enter the environment only when the predetermined set of criteria has been met; or at least one of: hold the vehicle in its current position when the predetermined set of criteria has not been met; and propose an alternative route for the vehicle when the predetermined set of criteria has not been met.

9. The method according to any preceding claim, further comprising: predicting at least one dynamic feature based on previously monitored dynamic features stored within a memory and/or external data; and modifying the baseline model to incorporate the at least one predicted dynamic feature.

10. The method according to claim 9, wherein the predicted dynamic feature and the monitored dynamic feature are the same dynamic feature.

11. The method according to claim 9 or claim 10, wherein the predicted dynamic feature is a dynamic feature expected to occur within the environment in the future.

12. The method according to any of claims 9 to 11, further comprising: updating the predicted dynamic features based on monitored dynamic features.

13. The method according to any preceding claim, further comprising: updating the baseline model to incorporate a subsequent environment; predicting at least one dynamic feature within the subsequent environment that is expected to occur when the vehicle is located within the subsequent environment; modifying the updated model to incorporate the at least one predicted dynamic feature within the subsequent environment; determining whether the modified updated model meets a predetermined set of criteria; and sending an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

14. A system for carrying out the method according to any preceding claim, wherein the system comprises: a memory comprising the base line model of the environment, a sensor configured to detect the electromagnetic waves emitted within the environment; a first processor configured to determine at least one parameter of at least one dynamic feature within the environment based on the electromagnetic waves detected by the sensor; a second processor configured to modify the baseline model such that it incorporates the at least one parameter of at least one dynamic feature; a third processor configured to determine whether the modified model meets a predetermined set of criteria; and a communication module configured to generate an output configured to allow the vehicle to enter the environment only when the predetermined set of criteria has been met.

15. The system according to claim 14, further comprising a transmitter configured to emit electromagnetic waves within the environment.

16. The system according to claim 14 or claim 15, wherein the first, second and third processors are configured to operate continuously in real-time or near real-time.

17. The system according to any one of claims 14 to 16, wherein the sensor configured to detect the electromagnetic waves emitted within the environment is ground based.