GB2538422A

GB2538422A - System and methods for predicting wave impacts with a watercraft, and watercraft control systems and methods

Info

Publication number: GB2538422A
Application number: GB1612579.1A
Authority: GB
Inventors: Munn Timothy; Gerald Stove Andrew
Original assignee: Thales Holdings UK PLC
Current assignee: Thales Holdings UK PLC
Priority date: 2014-06-10
Filing date: 2014-06-10
Publication date: 2016-11-16
Anticipated expiration: 2034-06-10
Also published as: GB2527055A; GB201410249D0; GB2538421B; GB201612577D0; GB2538421A; GB201612579D0; GB2527055B; GB2538422B

Abstract

A control system 100 for a watercraft 10 comprises a subsystem 110 for predicting wave impacts to the watercraft and a control unit 120. The subsystem for predicting wave impacts with the watercraft comprises a sensor (220, Fig 2) configured to detect signals reflected and /or scattered by waves on a water surface, and a processing unit (310, Fig 3) configured to generate wave prediction data indicating the predicted times of arrival of waves at the watercraft from the detected signals. The processing unit comprises a wave detection unit (330, Fig 3) and a prediction unit (340, Fig 3) that comprises a wave tracking unit (342, Fig 3) that uses an alpha-beta tracker to track the locations of waves over time. The control unit is configured to generate control signals for a throttle and/or a steering device 130 of the watercraft, in response to the wave prediction data, before the predicted time of arrival, to reduce the effect of the impact of the wave with the watercraft.

Description

Systems and methods for predicting wave impacts with a watercraft, and watercraft control systems and methods

FIELD

Embodiments of the present invention relate to control systems and wave impact prediction systems for watercrafts and in particular control systems and wave impact prediction systems for unmanned surface vehicles.

BACKGROUND

Unmanned surface vehicles (USVs) are watercrafts that operate on the water surface without an on-board crew. USVs have applications in the military and the oil and gas domains. A typical USV has a control system that sets the heading and speed of the watercraft. However, USV control systems do not currently take into account the prevailing sea and wave conditions into account. This can lead to damage to a USV caused by excessive slamming and the possibility of engine damage when the propellers leave the water.

In a manned vessel, the helmsman looks ahead and sets the heading and throttle of the vessel to minimise the impact to the vessel of waves and therefore reduce the possibility of damage occurring. Without the presence of a helmsman, current USVs are not able to make adjustments to minimise wave impacts.

SUMMARY OF THE INVENTION

According to a first aspect of the present invention a control system for a watercraft comprises a subsystem for predicting wave impacts to the watercraft and a control unit. The subsystem for predicting wave impacts with the watercraft comprises a sensor configured to detect signals reflected and/or scattered by waves on a water surface and to provide detection signals in response to the detected signals, and a processing unit configured to generate wave prediction data indicating predicted times of arrival of waves at the watercraft from the detected signals. The processing unit comprises a wave detection unit for determining locations of waves on the water surface relative to the watercraft from the detection signals and a prediction unit for generating the wave prediction data using the locations of the waves on the water surface relative to the watercraft. The prediction unit comprises a wave tracking unit for tracking the locations of the waves over time and the wave tracking unit comprises an alpha-beta tracker.

The control unit is configured to generate control signals for a throttle and/or a steering device of the watercraft, in response to the wave prediction data, before the predicted time of arrival, to reduce the effect of the impact of the wave with the watercraft.

The control system allows an unmanned surface vehicle (USV) to modify its navigation taking into account local wave situation. Thus the USV can react to predicted wave impacts as a competent skipper would and minimise the effect of the impact of the waves with the craft.

The sensor(s) is/are typically mounted on the front of the USV. Reflections are processed to determine the location of the oncoming sea waves and their severity. The wave information is then classified to determine which parts of the oncoming sea should be avoided and this information is passed to the USV control system that determines the most appropriate heading and throttle setting for the USV to successfully negotiate the oncoming sea without damage.

In an embodiment the sensor is further configured to provide detection signals in response to the detected signals. The processing unit comprises a wave detection unit for determining locations of waves on the water surface relative to the watercraft from the detection signals; and a prediction unit for generating the wave prediction data using the locations of the waves on the water surface relative to the watercraft.

In an embodiment the sensor is a radar sensor. In an embodiment the radar sensor is configured to detect signals in the frequency range 70GHz to 110GHz.

In an embodiment the radar sensor is a frequency modulated continuous wave (FMCW) radar.

Given the application area is maritime; a solid state, sealed sensor arrangement with no moving parts is advantageous. Automotive radar sensors are mounted in an exposed position at the front of the vehicle and are exposed to wet, high temperature, high salt, high vibration environments during their operational life. Thus a radar sensor designed for use as an automotive radar sensor may be used.

In an embodiment, the prediction unit comprises a wave tracking unit. The wave tracking unit tracks the locations of waves over time and from this tracking, an estimated time of arrival of the waves at the watercraft can be estimated. The tracker combines detections of the waves seen on successive 'looks' of the radar to estimate their speed of approach, based on how quickly the range changes and to smooth the estimate of the current position of the wave.

The tracking unit may be implemented as an alpha-beta tracker. The short observation time and possibly irregular motion of the wave fronts means that there is often no practical benefit in using a more complicated scheme. The alpha-beta tracker is then advantageous because of its low computation load. This type of tracker is described for example in 'Tracking and Kalman Filtering Made Easy' by E. Brookner, New York 1998, Wiley, ISBN 0-471-18407-1.

In an embodiment the wave detection unit is configured to compare a magnitude of the detection signals with a threshold and to determine the presence of a wave if the magnitude of a detection signal is greater than the threshold. The magnitude of the signals may be divided by a normalised value over a time period before comparison with the threshold.

In an embodiment the normalised value is the median value for the cell. In order that the 'average' (the normalised value) should not be influenced by the levels of the targets' (the returns from the sea) when they occur, the median is used in embodiments. The median is insensitive to extreme values, unlike the mean. Since the waves move in range over time, the return from a wave will only be present in a single range cell for a short time and so will have a negligible effect on the median value.

According to a second aspect of the present invention, a system for predicting wave impacts to a watercraft comprises a sensor, a wave detection unit and a wave prediction unit. The sensor is configured to detect signals reflected and /or scattered by waves on a water surface and to provide detection signals in response to the detected signals. The wave detection unit determines locations of waves on the water surface relative to the watercraft from the detection signals. The wave prediction unit generates wave prediction data using the locations of the waves on the water surface, the wave prediction data indicating, for at least one wave, a predicted time of arrival at the watercraft. The wave prediction unit comprises a wave tracking unit for tracking the locations of the waves over time and the wave tracking unit comprises an alpha-beta tracker According to a third aspect of the present invention there is provided a method of controlling a watercraft. The method comprises detecting signals reflected and/or scattered by waves on a water surface, determining locations of waves on the water surface relative to the watercraft, tracking the locations of the waves over time using an alpha-beta tracker, predicting times of arrival of waves at the watercraft from the detected signals using the tracked locations of the waves over time, and generating control signals for a throttle and/or a steering device of the watercraft, before the predicted time of arrival, to reduce the effect of the impact of the wave with the watercraft.

According to a fourth aspect of the present invention there is provided a method of predicting wave impacts to a watercraft. The method comprises detecting signals reflected and/or scattered by waves on a water surface and to providing detection signals in response to the detected signals; determining locations of waves on the water surface relative to the watercraft from the detection signals; tracking the locations of the waves over time using an alpha-beta tracker; and generating wave prediction data using the tracked locations of the waves over time, the wave prediction data indicating, for at least one wave, a predicted time of arrival at the watercraft.

A fifth aspect provides a computer program product comprising processor executable instructions which, when executed by a processor, cause the processor to perform a method as set out above. The computer program product may be embodied in a carrier medium, which may be a storage medium or a signal medium. A storage medium may include optical storage, or magnetic storage, or electronic storage.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following, embodiments are described, by way of example only, with reference to the accompanying drawings in which: Figure 1 shows a watercraft 10 which includes a control system 100 according to an embodiment of the present invention; Figure 2 shows a wave prediction system 210 according to an embodiment of the present invention; Figure 3 shows a wave prediction system according to an embodiment of the present invention; and Figure 4 shows a control system of a watercraft according to an embodiment of the present invention.

DETAILED DESCRIPTION

Figure 1 shows a watercraft 10 which includes a control system 100 according to an embodiment of the present invention. The watercraft 10 is an unmanned surface vehicle. The control system 100 generates control signals 125 for the throttle and rudder 130 of the watercraft 10. The control system 100 comprises a wave impact prediction subsystem 110 and a control unit 120. The wave prediction subsystem 110 generates wave prediction data 115 which indicates predicted times of arrival of waves at the watercraft 10. The control unit 120 generates the control signals 125 for the throttle and the rudder 130 of the watercraft 10 in response to the wave prediction data 115 to reduce the effects of the impact of the waves with the watercraft.

In embodiments, the wave prediction subsystem 110 comprises a sensor such as a radar sensor configured to detect signals reflected or scattered by waves on the water surface. One way to estimate the severity of the waves is from the strength of the signals reflected and / or scattered from the waves. From these signals, the wave prediction subsystem 110 generates wave prediction data 115 which indicates the time of arrival of individual waves. The operation of the wave prediction subsystem 110 is described in more detail below.

In addition to the time to arrival of waves at the watercraft, the wave prediction data 115 may also include indications of wave severity. The wave prediction data 115 may comprise a map of the wave field approaching the watercraft. In response to the wave prediction data 115, the control unit 120 may generate control signals to control the watercraft in a number of possible ways.

For example, the control unit 120 may generate control signals to cut the throttle to reduce the speed of the watercraft before the arrival of the wave at the watercraft. Alternatively, the control unit 120 may generate control signals to steer around oncoming waves. Further, the control unit may adjust the heading of the watercraft so the wave impacts with the front rather than the side of the watercraft. The control unit may generate signals to follow a path through the oncoming wave field which reduces the effect of the impact of the waves.

Figure 2 shows a wave prediction system 210 according to an embodiment of the present invention. The wave prediction system 210 may form the wave prediction subsystem 110 shown in Figure 1. The wave prediction system 210 comprises a sensor 220, a wave detection unit 230 and a prediction unit 240. In the embodiment described in detail below, the sensor 220 is a radar sensor, however, embodiments with other types of sensor are also envisaged. For example, a LIDAR or laser rangefinder is envisaged. A composite infra-red and visual sensor 3D such as a Kinect sensor is also envisaged, such sensors could be configured in pairs in order to extract range information.

The sensor 220 detects signals 215 reflected or scattered by waves on the water surface. In response to the detected signals, the sensor 220 generates detection signals 225 which indicate detected objects such as waves. The detection signals 225 may be indicative of a distance and a magnitude of the signal received from that distance. The wave detection unit 230 receives the detection signals and generates wave detection data 235 which indicates locations of waves on the water surface from the detection signals. The wave detection unit 230 differentiates between a target signal which represents a wave and other signals present which might come from noise in the sensor or signals from object present other than the target and interference. The wave detection unit may compare the detection signals with a threshold and if the signals are greater than the threshold output an indication that the detection signal corresponds to a wave on the water surface.

The wave prediction unit 240 receives the wave detection data 235 which indicates locations of waves on the water surface and generates wave impact prediction data 245 which indicates a predicted time of arrival of the waves at the watercraft.

In an embodiment, the wave prediction unit 240 receives wave detection data 235 which indicates locations of waves on the water surface for different times and tracks the location of the waves relative to the watercraft. From this tracking information, the wave impact prediction unit 240 determines a velocity of the waves relative to the watercraft. From the velocity of the waves and their locations, the wave prediction unit determines the time to arrival. Alternatively, the velocity of the waves relative to the watercraft may be determined by the wave detection unit 230, for example from Doppler signals detected by the sensor 220.

Figure 3 shows a wave prediction system according to an embodiment of the present invention. The wave prediction system 310 comprises a radar sensor 320, a wave detection unit 330 and a wave prediction unit 340.

The radar sensor 320 receives radar signals 315 and generates radar detection signals 325.

In one embodiment, the radar sensor 320 is a pulse radar sensor. In this case, the output of the radar sensor 320 is a series of detections indicating the ranges at which target echoes were detected for a particular occasion, for example after the transmission of a pulse. The range is indicated by the time between when a pulse is transmitted and when the pulse is received.

In an alternative embodiment, the radar sensor 320 is a frequency-modulated continuous wave (FMCW) radar sensor.

In the case of either a pulsed radar or FMCW radar, the output from the sensor 320 is a series of detections indicating the ranges at which target echoes were detected on a particular occasion, for example after the transmission of a pulse, the range being indicated by the time between when the pulse is transmitted and when it is received.

For a pulse radar, the processing chain within the radar sensor may be implemented as shown in Figure 1.4 of 'Introduction to Radar Systems,' by M. I. Skolnik, 3rd Ed., New York, 2001, McGraw-Hill, ISBN 0-07-290980-3, page 8, and the associated description.

Those of skill in the art will appreciate that the detection process is the same for both a pulsed radar and a frequency modulated continuous radar. A typical radar will transmit its signals many times so that it has many 'looks' in each direction within the period before it needs to make a decision about what targets are present. The sensitivity of the radar is typically enhanced by summing together the signals received from all these transmissions before the comparison with a threshold. Here the term 'transmissions' is used to cover both pulses and frequency modulation sweeps.

The wave detection unit 330 compares the magnitude of the signals with a threshold to determine between a 'target' and other signals present, which might come from receiver noise, returns from objects other than the intended targets (so-called clutter) and interference from other transmitters.

In an embodiment, since it was found that since the receiver noise and any interference extended over all bearings and was present over many looks of the system, a long-term average could be used to establish the level of the background in each range cell.

Creating a normalised value for each range cell independently makes the system insensitive to changes in receiver gain with range, which might be used in a practical radar to reduce the dynamic range of the signals (see for example 'Linear FMCW Radar Techniques,' A. G. Stove, IEE Proc F, Radar Signal Process., 1992, 139, (5), pp343-350).

In order that the normalised value should not be influenced by the levels of the 'targets' (the returns from the sea) when they occur, the 'average' used was the median, which is insensitive to extreme values, rather than the mean. Since the waves move in range over time, the return from a wave will only be present in a single range cell for a short time and so will have on a negligible effect on the median value.

The signal level in each range cell was divided by its normalised value, after which the residual background was just the variance of the thermal noise, and the targets could be discriminated from that using a simple fixed threshold.

Note that the system must typically judge how severe the wave is to determine whether or not it should avoid it. The strength of the radar return as a multiple of the threshold value is used for this purpose and so is reported at the output of the threshold decision as well as the range.

Since the crest of the wave, from which most of the reflection comes, may be more than a metre in length whereas the range cell of the radar will generally be less than one metre, the return from a single wave will occupy many contiguous range cells.

These successive returns are merged into a single report giving the range and amplitude of the biggest signal within the group.

In the embodiment shown in Figure 3, wave prediction unit comprises a wave tracking unit 342. This combines detections of the waves seen on successive 'looks' of the radar to estimate their speed of approach, based on how quickly the range changes and hence to smooth the estimate of the current position of the wave.

In an embodiment, the wave tracking unit 342 is implemented as an alpha-beta tracker.

The alpha-beta tracker is preferred because of its much simpler computation load.

Whilst many other more sophisticated tracker algorithms are known, the short observation time and possibly irregular motion of the wave fronts means that there is often little practical benefit in using a more complicated scheme.

Once a track is established it will be passed to the control system only once a number of detections have been seen on the same wave. At least two detections are needed to measure the closing speed and the tracker design may demand that these two detections are seen within three successive looks' by the radar to minimise the possibility of a false track being established on noise or interference.

Once established, a track can be 'dead reckoned' i.e. the estimated position can be updated using knowledge of its previous position and its speed, even if some detections are missing. This may in fact happen at short range when the wave may be hidden from view from the radar by the structure of the watercraft, but useful information can still be given to the control system. The track will typically be 'deleted' when the estimated time to impact has passed.

In many implementations, the radar sensor may survey a relatively narrow angular field of view (for example 20 degrees) with a relatively broad resolution (for example five degrees). In this situation it is possible to track the waves only in range, i.e. treat each angular sector independently.

The output from the wave prediction unit 340 is the wave prediction data 345 which comprises the time to impact, i.e. the time before the wave hits the boat. This is derived simply by dividing the range by the approach velocity (the rate of reduction of range with time).

The time to impact for all the waves whose returns are above the level where they might cause a problem to the boat may typically be put into a time 'queue' so the boat's controller will react first to that which will hit the boat first.

Figure 4 shows a control system of a watercraft according to an embodiment of the present invention. Waves 402 on a water surface reflect radar signals 404 emitted by a radar module 406. The radar module may be implemented as an automotive radar sensor. Automotive radar sensors typically operate in the range 76 to 77 GHz. There are a number of advantages to using an automotive radar sensor. Automotive radar sensors generally operate as FMCW radars and are solid state devices with no moving parts. They have small size and low power requirements. They are packaged to be resistant to a harsh environment, for example they are generally supplied in waterproof housing which is resistant to vibrations.

Detection outputs 408 from the radar 406 are received by a data processing module 410. The data processing module includes the wave detection unit and the wave impact prediction unit described above. The output of the data processing module is wave prediction data 412 which indicates predicted times of arrival of waves. The wave prediction data 412 may also indicate positions of waves.

The wave prediction data 414 is input into the control module 414 of the watercraft. The control module generates control signals 416 and 418 for the throttle 422 and the rudder 420 of the watercraft.

The control systems and wave impact prediction systems described above may be implemented as software running on any suitable hardware for example a computer running Linux. The computer may take any form such as PC104 format, laptop, industrial 19" rack PC or Raspberry Pi. Alternatively, system may be implemented as hardware such as a field programmable gate array, or as a combination of hardware and software. The described embodiments can be incorporated into a specific hardware device, a general purpose device configured by suitable software, or a combination of both. Aspects can be embodied in a software product, either as a complete software implementation, or as an add-on component for modification or enhancement of existing software (such as a plug in). Such a software product could be embodied in a carrier medium, such as a storage medium (e.g. an optical disk or a mass storage memory such as a FLASH memory) or a signal medium (such as a download). Specific hardware devices suitable for the embodiment could include an application specific device such as an ASIC, an FPGA or a DSP, or other dedicated functional hardware.

Claims

CLAIMS: 1. A control system for a watercraft, the control system comprising a subsystem for predicting wave impacts to the watercraft, comprising: a sensor configured to detect signals reflected and/or scattered by waves on a water surface and to provide detection signals in response to the detected signals, a processing unit configured to generate wave prediction data indicating predicted times of arrival of waves at the watercraft from the detected signals, the processing unit comprising: a wave detection unit for determining locations of waves on the water surface relative to the watercraft from the detection signals; and a prediction unit for generating the wave prediction data using the locations of the waves on the water surface relative to the watercraft, wherein the prediction unit comprises a wave tracking unit for tracking the locations of the waves over time and wherein the wave tracking unit comprises an alpha-beta tracker; and a control unit for generating control signals for a throttle and/or a steering device of the watercraft, in response to the wave prediction data, before the predicted time of arrival, to reduce the effect of the impact of the wave with the watercraft.
2. A control system for a watercraft according to claim 1 wherein the sensor is a radar sensor.
3. A control system for a watercraft according to claim 2 wherein the radar sensor is configured to detect radar signals in the frequency range 70GHz to 110GHz.
4. A control system for a watercraft according to claim 2 or claim 3, wherein the radar sensor is a frequency modulated continuous wave radar.
5. A control system according to any preceding claim wherein the subsystem for predicting wave impacts is configured to output an indication of the potential severity of the impact of the wave with the watercraft.
6. A control system according to claim 5, wherein the wave detection unit is configured to determine the indication of the potential severity of the impact of the wave with the watercraft from the strength of the detected signals.
7. A control system according to any preceding claim, wherein the wave detection unit is configured to compare a magnitude of the detection signals with a threshold and to determine the presence of a wave if the magnitude of a detection signal is greater than the threshold.
8. A control system according to claim 7, wherein the wave detection unit is configured to divide the magnitude of the detection signals by a normalised value over a time period for a range cell of the sensor then compare the result with a threshold.
9. A control system according to claim 8, wherein the normalised value is the median value.
10. A system for predicting wave impacts to a watercraft, the system comprising a sensor configured to detect signals reflected and/or scattered by waves on a water surface and to provide detection signals in response to the detected signals; a wave detection unit for determining locations of waves on the water surface relative to the watercraft from the detection signals; and a wave prediction unit for generating wave prediction data using the locations of the waves on the water surface, the wave prediction data indicating, for at least one wave, a predicted time of arrival at the watercraft, wherein the wave prediction unit comprises a wave tracking unit for tracking the locations of the waves over time and wherein the wave tracking unit comprises an alpha-beta tracker.
11. A system for predicting wave impacts to a watercraft according to claim 10, wherein the sensor is a radar sensor.
12. A system for predicting wave impacts to a watercraft according to claim 11 wherein the radar operates in the frequency range 70GHz to 110GHz.
13. A system for predicting wave impacts to a watercraft according to any one of claims 10 to 12 further configured to output an indication of the potential severity of the impact of the wave with the watercraft.
14. A system for predicting wave impacts to a watercraft according to claim 13, wherein the wave detection unit is configured to determine the indication of the potential severity of the impact of the wave with the watercraft from the strength of the detected signals.
15. A system for predicting wave impacts to a watercraft according to any one of claims 10 to 14, wherein the wave detection unit is configured to compare a magnitude of the detection signals with a threshold and to determine the presence of a wave if the magnitude of a detection signal is greater than the threshold.
16. A system for predicting wave impacts to a watercraft according to claim 15, wherein the threshold is a normalised value over a time period for a range cell of the sensor multiplied by a factor selected to give an acceptable balance between sensitivity and false alarm rate.
17. A system for predicting wave impacts to a watercraft according to claim 16, wherein the normalised value is the median value.
18. A method of controlling a watercraft, the method comprising detecting signals reflected and/or scattered by waves on a water surface; determining locations of waves on the water surface relative to the watercraft; tracking the locations of the waves over time using an alpha-beta tracker; predicting times of arrival of waves at the watercraft from the detected signals using the tracked locations of the waves over time; and generating control signals for a throttle and/or a steering device of the watercraft, before the predicted time of arrival, to reduce the effect of the impact of the wave with the watercraft.
19. A method according to claim 18 wherein the signals reflected and/or scattered by waves on a water surface are radar signals.
20. A method according to claim 19 wherein the radar signals are in the frequency range 70GHz to 110GHz.
21. A method according to one of claims 18 to 20 further comprising comparing a magnitude of the detection signals with a threshold and to determine the presence of a wave if the magnitude of a detection signal is greater than the threshold.
22. A method according to claim 21 further comprising dividing the magnitude of the detection signals by a normalised value over a time period for a range cell of the sensor then comparing the result with the threshold.
23. A method according to claim 22, wherein the normalised value is the median value.
24. A method for predicting wave impacts to a watercraft, the method comprising detecting signals reflected and/or scattered by waves on a water surface and to providing detection signals in response to the detected signals; determining locations of waves on the water surface relative to the watercraft from the detection signals; tracking the locations of the waves over time using an alpha-beta tracker; and generating wave prediction data using the tracked locations of the waves over time, the wave prediction data indicating, for at least one wave, a predicted time of arrival at the watercraft.
25. A method according to claim 24 wherein the signals reflected and/or scattered by waves on a water surface are radar signals.
26. A method according to claim 25 wherein the radar signals are in the frequency range 70GHz to 110GHz.
27. A method according to one of claims 24 to 26 further comprising comparing a magnitude of the detection signals with a threshold and to determine the presence of a wave if the magnitude of a detection signal is greater than the threshold.
28. A method according to claim 27 further comprising dividing the magnitude of the detection signals by a normalised value over a time period for a range cell of the sensor then comparing the result with the threshold.
29. A method according to claim 28, wherein the normalised value is the median value.
30. A computer readable carrier medium carrying processor executable instructions which when executed on a processor cause the processor to carry out a method according to any one of claims 18 to 29.