AU658412B2 - Optimising radar tasking using backwards scheduling - Google Patents

Description

40555AU

HKS:MAH:LL

P/00/011 Regulation 3 2

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

Name of Applicant: *.THE COMMONWEALTH OF AUSTRALIA SActual Inventors: GEORGE FREDERICK EARL DOUGLAS JOHN KEWLEY BRUCE DONALD WARD Address for Service: SCOLLISON CO.,117 King William Street, Adelaide, S.A. 5000 Invention Title: OPTIMISING RADAR TASKING USING BACKWARDS SCHEDULING Details of Associated Provisional Applications: Patent Application No. PK63000 dated 24th May 1991 The following statement is a full description of this invention, including the best method of performing it known to us: BACKGROUND OF THE INVENTION The invention described herein relates to radar tasking for over-the-horizon radar systems. It is envisaged however that the invention may be applicable to other high frequency radio based systems.

Over-the-horizon radar and other high frequency radio systems make use of ionospheric propagation. Ionospheric propagation allows radio waves to propagate large distances. Unlike microwave radio systems a transmitter and a receiver do not need to be within line of sight. Ionospheric propagation, whilst at times allowing for very long distance radio signal propagation, at other times allows only very much shorter radio wave propagation. This phenomena is well known. Therefore a particular task required of an overthe-horizon radar or another high frequency radio based system may or may not be successfully completed at any particular moment. So it is desirable to know the likelihood or probabilityof successful completion of a required task before attempting the required task. It is to this problem that the invention is directed.

The problems to which this invention is directed are not entirely deterministic.

•Whilst physical characteristics can be estimated, such as the height of the E and F iayers, other parameters are not measurable or determinable. Further, 20 some predictions about successful propagation of radio waves through the ionosphere depend upon a large number of variables and require complex S:i. computations to be conducted for a deterministic result to be obtained. Such predictions can often be approximated by heuristic reasoning. To be useful as a task scheduling tool the prediction needs to be completed quikly, that is in real time.

Using an over-the-horizon radar system as an example, such systems are generally run to do specifically requested tasks immediately the task is requested. This means that tasks are not always conducted when the •ionospheric conditions are suitable for successful completion of those tasks.

Further, changing from one task to another requires the radar system parameters to be changed. These parameters can depend on the type of task requested and the current characteristics of the ionosphere. So changing Stasks can be a complex and difficult procedure requiring extensive training 3 and intuitive feel for the radar system. These problems conspire to make this known manner of operating these radar systems less efficient than is desired.

Known methods of operating an over-the-horizon radar system generally require the operators to have a broad understanding in a wide range of disciplines. These are in addition to skills required to determine radar returns.

The training of such operators is a long process and often requires a team of people to operate the radar.

In the case of over-the-horizon radar systems it is an intended object of the invention to reduce the number of operators and the degree of skill required to operate a radar system by providing means which will automatically determine the operating parameters of the radar system Furthermore, such means will assist in scheduling tasks by providing a measure of likely successful completion of a requested task. The measurable parameters needed for this means are generally available to radar systems of this type.

It is an intended object of the invention to provide means for predicting successful task performance based upon measured parameters and heuristic reasoning. Such a means can operate in reai time and provide system operating settings most likely to achieve the successful completion of a requested task.

20 SUMMARY OF THE INVENTION The invention may be said to reside in a method to aid in the selection of radar tasks from a list of tasks, to optimise the use of an over-the-horizon radar means, comprising the steps of: predicting, the probability of successful completion of each of 25 said radar tasks, based upon ionospheric conditions, task requirements and S.heuristic reasoning; ranking said tasks in order of said probability; selecting the task with the highest rank; and causing the operating parameters of said radar meanis to be set to the parameters associated with said selected task, wherein the method is performed by computing means.

Preferably, the list of tasks includes proactive tasks and reactive tasks.

Preferably, the probability of successful completion of a task comprises: analysing current ionospheric conditions; (ii) assessing reactive tasks for probability of successful completion; and (iii) assessing proactive tasks for probability of successful completion.

i~referably, the tasks comprises: generating a Current Task List of tasks likely to be successfully completed given current ionospheric conditions in order of their probability of successful completion; (ii) determining the optimum operating parameters of the radar means to successfully complete the tasks in the Current Task List; and (iii) generating a schedule of the ordered tasks and their corresponding parameters to be applied to the radar means.

Preferably, the proactive tasks are maintained in the Proactive Task List in a memory means and periodically analysed for probability of successful completion.

Preferably, the step of determining the optimum operating parameters is •"performed by a computing means programmed with heuristic rules, and having as input prevailing ionospheric conditions and task requirements.

S

Preferably, a list of tasks are assessed and ranked in order of probability of successful completinn. If the probability of successful completion is not great the task can be postponed to a later more opportune time. In the case of radar tasking, a list of routine tasks can be prioritised in order of probability ol 25 successful completion and compared with a list of requested tasks.

Requested tasks which are not required to be done immediately or have a low probability can be postponed until the probability has improved. When the task is selected then the required operating parameters of the radar such as carrier frequency and waveform repetition rate, can be set automatically.

.IOO.i The method can be used to give the probability of successful completion of any task upon request. For example, if a radar operator wishes to detect targets within a certain area then the method will use the current prevailing conditions and heuristic information to assess the probability of successful completion of that task.

A computing means is used to assess the probability of successful completion of any task and a visual display means can display the result. Such a computing means would have as input the current conditions prevailing in the ionosphere. Such conditions can be soundings of the ionospheric E and F layer heights. The computer means provides an assessment of the probability of the successful completion of a task based upon heuristic rules and measured prevailing conditions. This assessment is then displayed upon a screen with the associated task. The task can be one of a number of tasks which are ordered in priority order of highest successful completion. The tasks and the order of priority are periodical!y updated.

In preference the computing means determines the appropriate operating parameters for the radar means directs the radar means to operate with these parameters. Through this invention a radar or high frequency radio based system may operate independently from external control such as normally provided by a skilled operator.

In preference there is provided memory means which contain a list of routine *oco tasks. Routine radar tasks are completed automatically unless specdfically 20 overridden. These routine tasks are commenced when the task has a good probability of being successfully conducted, as assessed by the computing means. Radar parameters are automatically set as required to do a particular task.

S

is expected for example, that a radar, when not conducting specifically requested tasks, can be conducting routine tasks automatically. By such means a radar can be used more efficiently thain when the radar is only operated to do specific requested tasks.

In a further form of the invention there is provided a radar operating means to ooo optimise the operation of an over-the-horizon radar means, comprising a task selection processing means, a frequency management processing means, and a task scheduler processing means, wherein: the task selection processing means predicts the probability of successful completion of each task in a list of radar tasks, based upon 6 ionospheric conditions, task requirements and heuristic reasoning, then generates a Current Task List of tasks in order of their predicted probability of successful task completion; the frequency management processing means provides to the task scheduler processing means the optimum radar operating para~meters to be applied to the radar means; and the task scheduler processing means generates a schedule of the ordered tasks and their corresponding radar operating parameters, selects the task with the highest rank and then causes the operating parameters corresponding to this task to be applied to the radar means.

In a further form of the invention there is provided a task assessment processing means, which compares actual task performance with predicted task performance and provides feedback to the task selection processing means to update the Current Task List.

In a further form of the invention there is provided the lists of tasks includes proactive tasks and reactive tasks.

In a further form of the invention there is provided display means to display the Current Task List, predicted task performance indicia, actual task performance indicia, and the results of completed tasks.

DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment will now be described with reference to a drawing in which il FIG. 1 is a schematic of a processing means in accordance with the ti invention.

25 The following embodiment is described in reference to an over-the-horizon radar system operating at radio frequencies in the high frequency band although it will be appreciated that other high frequency radio systems can use the invention in other specific embodiments. The described embodiment is for illustrative purposes only and is not to be taken as limiting of the invention.

It will be appreciated that the ultimate performance achieved by a sophisticated system such as an over-the-horizon radar is very dependent on the skill level and experience of the operator. If every operator were an expert the quality of surveillance missions would be greatly enhanced. Since this is not generally possible the benefit of expert knowledge is built in to a system for automatically operating the radar.

FIG. 1 shows an operationall flowchart consisting of a number of inter-related processes each comprised of a number of modules and data flows. Each item in the flowchart is described below.

Task Selection Process I The task selection process consists of the Task Parameter Selection module and Radcon Interface module. The Radcon interface module manages the operator interface and controls much of the operation of the rest of the system. It receives task requests from external agencies and compiles these into the New Task List.

The Proactive Task List is a list of tasks vwiti.g to be performed on a proactive basis. The Radcon Interface module periodically passes these into the New Task List. The New Task List is a list of reactive tasks received from request agencies as well as proactive tasks to be considered for inclusion to the Current Task List.

20 The Task Parameter Selection module accepts data from the New Task List and produces an optimal radar configuration for each task in such a way as to maximise overall radar performance rather than optimizing each task separately. The module communicates with the FMS Advice module in :determining the optimum radar configuration and calculates Expected 25 Performance Indices which are provided to the Radcon Interface with the task list. The Radcon Interface generates or modifies a Current Task List based on the performance indices, priorities and rule based decision algorithms.

The Proactive Task List contains a list of about 20 generic tasks that will be run in the absence of higher priority tasks. These tasks include calibration tasks as well as low priority tasks in "an area of interest" such as general coastline surveillance. Each proactive task will be characterised by predefined mission type predefined target type predefined range predefined range depth predefined azimuth range predefined revisit time priority Each reactive task will have the following characteristics mission type target type range range depth azimuth range priority revisit time starting time preferred completion time The requesting agency may not have a priori assumptions about the S°environmental condition, radar configuration or waveform parameters which o 20 are best suited to the mission. Since the performance of a radar is dependent upon prevailing propagation conditions, there can be no guarantee that a mission requested from an external agency can be achieved immediately.

Likewise, there is a possibility that the radar could adequately perform a mission which has not been requested but which, if brought to the attention of 25 the tasking agency, may be recognised as meeting a need. Thus a combination of the reactive and proactive lists will be the most appropriate radar tasking schedule at any given time. The chosen combination becomes the Current Task List.

Frequency Management Process 2 Current Ionospheric Conditions are 30 continually monitored by a Frequency Management System (FMS) and a FMS Database is maintained. The FMS Advice module analyses FMS Data in response to requests for frequency advice from the Task Performance "Selection module or the Current Task Performance Analyzer module. The FMS Advice module uses an algorithmic approach to calculate the optimum operating parameters for the radar at a given time for a given radar configuration.

Task Scheduler Process 3 The Scheduler process consists of the Scheduler module which schedules tasks for the radar based upon the tasks and priorities contained in the Current Task List and constraints provided from the Radcon Interface. The schedule determined by the Scheduler is passed to the receive and transmit sites of the radar and is displayed on display means.

Task Assessment Process 4: The Assessment process monitors task performance and provides feedback for task update. The Current Task Performance Analyzer module takes the Current Task List and calculates the Task Performance Indices for each task on the list. It calculates both the expected performance by interfacing with the FMS Advice module and the real performance by looking at the data in the Current Radar Performance Database.

These indices are passed to the Task Parameter Updater module which decides which tasks need to have parameters changed to improve performance. The Task Parameter Updater module communicates these .changes to the Radcon Interface to close the feedback loop.

0 In summary the data-flow into the processing system consists of task requests S 20 from external agencies, ionospheric conditions and current radar operations.

The data out-flow is the radar operating parameters. All decisions relating to the optimum operation of the radar occur within the system.

The modular structure of the invention is extended to an object oriented approach. The object oriented representations are unique in that all objects communicate with one another by sending and receiving messages. When an object receives a message, it consults its database and rules to decide what 00., action to take. The rules may be stored directly with the object or in a higherlevel object in the network hierarchy.

000000 The invention can be applied to radar systems or other high frequency systems with one or more stations. In the case of a radar network, radars may Sbe task scheduled with use of the invention.

It will be appreciated that there are many ways in which to conduct the invention disclosed herein and all such embodiments of the invention would fall within the spirit of the invention.

e e*•

*S

*e

Claims (9)

1. A method to aid in the selection of radar tasks from a list of tasks, to optimise th3 use of an over-the-horizon radar means, comprising the steps of: predicting the probability of successful completion of each of said radar tasks, based upon ionospheric conditions, task requirements and heuristic reasoning; ranking said tasks in order of said probability; selecting the task with the highest rank; and causing the operating parameters of said radar means to be set to the parameters associated with said selected task, wherein the method is performed by computing means.

2. The method of claim 1 in which the list of tasks includes proactive tasks and reactive tasks.

3. The method of claim 2 wherein the step of predicting the probability of successful completion of a task comprises: analysing current ionospheric conditions; (ii) assessing reactive tasks for probability of successful completion; and assessing proactive tasks for probability of successful 20 completion.

4. The method of claim 1 wherein the step of ranking the tasks comprises: generating a Current Task List of tasks likely to be successfully completed given current ionospheric conditions in order of their probability of successful completion; (ii) determining the optimum operating parameters of the radar means to successfully complete the tasks in the Current Task List; and (iii) generating a schedule of the ordered tasks and their corresponding parameters to !bs applied to the radar means. The method of either of claims 2 or 3 in which the proactive tasks are maintained in the Proactive Task List in a memory means and periodically analysed for probability of successful completion.

6. The method of claim 4 in which the step of determining the optimum operating parameters is performed by a computing means programmed with heuristic rules, and having as input prevailing ionospheric conditions and task requirements,

7. A radar operating means to optimise the operation of an over-the- horizon radar means, comprising a task selection processing means, a frequency management processing means, and a task scheduler processing means, wherein: the task selection processing means predicts the probability of successful completion of each task in a list of radar tasks, based upon ionospheric conditions, task requirements and heuristic reasoning, then generates a Current Task List of tasks in order of their predicted probability of successful task completion; the frequency management processing means provides to the task scheduler processing means the optimum radar operating parameters to be applied to the radar means; and the task scheduler processing means generates a schedule of the ordered tasks and their corresponding radar operating parameters, selects the task with the highest rank and then causes the operating parameters corresponding to this task to be applied to the radar means. A radar operating means as in claim 7, further comprising a task S: assessment processing means, which compares actual task performance with predicted task performance and provides feedback to the task selection E processing means to update the Current Task List.

9. The radar operating means of claim 7 in which the lists of tasks includes proactive tasks and reactive tasks. .9 A radar operating means as in claim 7, further including display means to display the Current Task List, predicted task performance indicia, actual task performance indicia, and the results of completed tasks. oo •I

11. A method of operating a radar means as herein described with reference to the accompanying drawings.

12. A radar operating means as herein described with reference to the accompanying drawing. Dated this 10th day of January 1995 THE COMMONWEALTH OF AUSTRALIA By i3Patent Attorneys, COLLISON CO. so. OV 4 ABSTRACT A method of predicting successful task performance based upon measured parameters and heuristic reasoning including the steps of predicting successful operation of terminal radio tasks based upon ionospheric conditions, task requirements and heuristic information, and causing the operating parameters of radar means to be set to parameters associated with the task prediction. A list of tasks are assessed and ranked in order to likely successful completion. The likelihood of successful task completion is determined by inputting into a computing means the current prevailing conditions in the ionosphere and other parameters defining the radar task. The computing means is programmed with heuristic rules which establish the likelihood of completion. Tasks are then prioritised and listed in order of likely successful completion. S o* S