Disclosure of Invention
Solves the technical problem
Aiming at the defects of the prior art, the invention provides a sensor signal redundancy management method for a large-scale freight transport unmanned aerial vehicle.
Technical scheme
In order to solve the above problems, the present invention provides the following technical solutions: a large-scale freight transport unmanned aerial vehicle sensor signal redundancy management method, the flight control computer monitors and votes the analog input signal that the redundancy sensor outputs, the management of sensor redundancy includes the monitoring and votes of the four redundancy signal, the monitoring and votes of the three redundancy signal, the monitoring and votes of the two redundancy signal. The specific method comprises the following steps: data initialization → data synchronization → data acquisition → input data cross transmission → input data comparison → data fusion → control law calculation → output data cross transmission → output data comparison → data transmission → data voting circuit → airborne terminal → fault handling;
(1) monitoring and voting of quad redundancy signals
Firstly, ordering the four redundancy signals, then calculating differences in pairs, comparing the difference value with a given threshold value, taking a majority for voting, and when the four signals have uncertain faults, taking the on-line monitoring of the sensor as a reference, and taking the signal which is considered to be correct by the on-line monitoring as a normal signal;
(2) monitoring and voting of three-redundancy signals
When the four-redundancy signal has a one-time permanent fault and the one-time fault is not recovered, the four-redundancy sensor is reduced to a three-redundancy sensor to continue working, at the moment, the three-redundancy voting monitoring is used for processing, the three-redundancy signals are sequenced, the median value is subtracted from the other two signals respectively, the difference value is compared with a given tolerance value, and when the uncertain fault occurs, the on-line monitoring of the sensor is used as the reference, and the on-line monitoring is used for considering the correct signal to be a normal signal;
(3) monitoring and voting of two redundancy signals
When the triple redundancy signal has a permanent fault again, the triple redundancy sensor is reduced to a double redundancy sensor to continue working, the double redundancy voting monitoring principle is used for processing, the difference of the double redundancy signals is calculated, the difference value is compared with a given threshold value, if the difference value exceeds the threshold value, the on-line monitoring of the sensor is taken as a reference, the on-line monitoring considers that the correct signal is a normal signal, the double redundancy signals are voted, and the average value is taken as the output of the voting value.
Preferably, when the input data and the output data are transmitted in a cross manner, a data cross link algorithm is adopted, the flight control computer disassembles the calculated output instruction into data frames, the data frames are sequentially transmitted, and after the receiving node receives the data frames, the data information in the data frames is taken out and assembled into data.
Preferably, when the flight control computer performs redundancy management, a task-level synchronization algorithm is used to compare and vote the calculation results, and the specific steps are as follows:
s1, in each operation period, firstly operating an upper application program, performing flight control calculation, and entering a synchronization point after a calculation result is obtained;
s2 at the synchronization point, the flight control computer first generates the synchronization frame of the period, carries the record and sends the serial number of the flight control computer, runs the period serial number information, and broadcasts to other flight control computers;
s3, the flight control computer checks the received synchronization frame of other flight control computers, and judges that the other flight control computers enter the synchronization point or are in a synchronization receiving waiting state;
s4 receiving the synchronous frame of other flight control computers and judging the related information to judge whether the synchronous frame is reasonable;
s5, if the flight control computer receives reasonable synchronous frames of other flight control computers before the synchronous waiting time reaches the upper limit, the synchronization is successful, the time between the flight control computer entering the synchronous point and the last synchronous frame is calculated, the value is used as a time difference parameter to be transmitted to upper application software, the synchronization is finished, otherwise, the synchronization of the flight control computer is considered to be failed;
s6 informs the upper layer application program to delay according to the time difference parameter delivered by the synchronization algorithm to ensure that the local flight control computer and the application program of the slowest flight control computer in the network start simultaneously in the next operating cycle.
Preferably, the redundancy management method is realized by a plurality of independent tasks, each flight control cycle is realized by sending an operation result to the shared memory after the flight control software completes operation, then sending a message to the redundancy management interface and starting the task, the redundancy management interface sequentially starts a synchronous task, receives/sends bus tasks to process data in the shared memory, updates a system view according to the processing results of the tasks, calculates the membership relationship and carries out data voting, and finally the redundancy management interface task returns the operation result of the operation cycle to the upper application software.
Preferably, a voting algorithm is adopted to shield the redundancy flight control computer system from faults, and based on vector distance calculation, the voting algorithm is as follows:
(1) calculating the distance between the packaging expression vectors of any two flight control computers;
(2) if the distance between the packaging representation vectors of the two flight control computers is smaller than the threshold value determined by the user, respectively casting a ticket to the two flight control computers;
(3) after the data of all the flight control computers are processed, the ticket obtaining conditions of all the flight control computers are counted, the flight control computers with most tickets exceeding 2/3 are normal, the flight control computers with less tickets than 1/3 are abnormal, and the final output result is the resolving result of the flight control computer with the largest ticket number.
Advantageous effects
The invention provides a large-scale freight unmanned aerial vehicle sensor signal redundancy management method, which has the following beneficial effects:
the invention can increase redundancy resources and shield the influence of fault parts through signal monitoring and voting of four redundancies based on a task-level synchronization algorithm and a voting algorithm, thereby improving the task reliability and flight reliability of the flight control computer.
Detailed Description
All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Further explaining a technical scheme provided by the invention:
a large-scale freight transport unmanned aerial vehicle sensor signal redundancy management method, the flight control computer monitors and votes the analog input signal that the redundancy sensor outputs, the management of sensor redundancy includes the monitoring and votes of the four redundancy signal, the monitoring and votes of the three redundancy signal, the monitoring and votes of the two redundancy signal. The specific method comprises the following steps: data initialization → data synchronization → data acquisition → input data cross transmission → input data comparison → data fusion → control law calculation → output data cross transmission → output data comparison → data transmission → data voting circuit → airborne terminal → fault handling;
(1) monitoring and voting of quad redundancy signals
Firstly, ordering the four-redundancy signals, then carrying out pairwise difference calculation, comparing the difference value with a given threshold value, taking a majority for voting, when the four signals have uncertain faults (such as 2:2 faults), taking the online monitoring of a sensor as a reference, taking the signal which is considered to be correct by the online monitoring as a normal signal, and taking the average value with the largest degree as the output of a voting value when the four-redundancy signals are voted;
(2) monitoring and voting of three-redundancy signals
When the four-redundancy signal has a one-time permanent fault and the one-time fault is not recovered, the four-redundancy sensor is reduced to a three-redundancy sensor to continue working, then the three-redundancy voting monitoring is used for processing, the three-redundancy signals are sorted, the median value is subtracted from the other two signals respectively, the difference value is compared with a given tolerance value, similarly, when an uncertain fault (1:1:1 fault) occurs, the on-line monitoring of the sensor is used as the reference, the on-line monitoring is used for judging that the correct signal is a normal signal, and when the three-redundancy signal votes, the median value is used as the voting value to be output;
(3) monitoring and voting of two redundancy signals
When the triple redundancy signal has a permanent fault again, the triple redundancy sensor is reduced to a double redundancy sensor to continue working, the double redundancy voting monitoring principle is used for processing, the difference of the double redundancy signals is calculated, the difference value is compared with a given threshold value, if the difference value exceeds the threshold value, the on-line monitoring of the sensor is taken as a reference, the on-line monitoring considers that the correct signal is a normal signal, the double redundancy signals are voted, and the average value is taken as the output of the voting value.
When the input data and the output data are transmitted in a cross mode, a data cross link algorithm is adopted, the flight control computer disassembles the output instruction obtained by calculation into data frames, the data frames are sequentially sent, and after the receiving node receives the data frames, the data information in the data frames is taken out and assembled into data. In order to ensure the integrity of the transmission process, an end frame is added as a mark at the end of the data frame, then sending according to the sequence of the data frame and the end frame, the end frame triggers the receiving node to check the data and the integrity, the sending node waits for the feedback of the receiving node, if there is a datagram frame loss, the lost datagram frame is retransmitted, the node receives datagrams and end frames of other nodes, if it is the end frame, it indicates that the sending node has finished sending the group data frame, the receiving node starts scanning the buffer area, checks whether all data frames are received, if there is a lost frame, the receiving node generates a retransmission request frame, requests the counterpart transmission bus to retransmit the lost data frame, if no frame is lost, the receiving node generates a successful confirmation frame and sends the successful confirmation frame to the transmitting node of the opposite side to inform the successful transmission.
When the flight control computer performs redundancy management, a task-level synchronization algorithm is used to compare and vote the calculation results, and the method specifically comprises the following steps:
s1, in each operation period, firstly operating an upper application program, performing flight control calculation, and entering a synchronization point after a calculation result is obtained;
s2 at the synchronization point, the flight control computer first generates the synchronization frame of the period, carries the record and sends the serial number of the flight control computer, runs the period serial number information, and broadcasts to other flight control computers;
s3, the flight control computer checks the received synchronous frame of other flight control computers, judges that other flight control computers enter a synchronous point or are in a synchronous receiving waiting state, in order to prevent the communication fault or the flight control computer fault from causing deadlock, the synchronous receiving waiting time must have an upper limit, and once the upper limit is reached, the synchronization is forcibly removed;
s4 receiving the synchronous frame of other flight control computers and judging the related information to judge whether the synchronous frame is reasonable;
s5, if the flight control computer receives reasonable synchronous frames of other flight control computers before the synchronous waiting time reaches the upper limit, the synchronization is successful, the time between the flight control computer entering the synchronous point and the last synchronous frame is calculated, the value is used as a time difference parameter to be transmitted to upper application software, the synchronization is finished, otherwise, the synchronization of the flight control computer is considered to be failed;
s6 informs the upper layer application program to delay according to the time difference parameter delivered by the synchronization algorithm to ensure that the local flight control computer and the application program of the slowest flight control computer in the network start simultaneously in the next operating cycle.
The redundancy management method is realized through a plurality of independent tasks, each flight control period is realized through the steps that after the flight control software finishes operation, an operation result is sent to a shared memory, then a message is sent to a redundancy management interface, the task is started, the redundancy management interface sequentially starts a synchronous task, receives/sends bus tasks to process data in the shared memory, updates a system view according to the processing results of the tasks, calculates the membership relation and carries out data voting, and finally the redundancy management interface task returns the operation result of the operation period to upper-layer application software.
Adopting a voting algorithm to shield the redundancy flight control computer system from faults, and based on vector distance calculation, the voting algorithm is as follows:
(1) calculating the distance between the packaging expression vectors of any two flight control computers;
(2) if the distance between the packaging representation vectors of the two flight control computers is smaller than the threshold value determined by the user, respectively casting a ticket to the two flight control computers;
(3) after the data of all the flight control computers are processed, the ticket obtaining conditions of all the flight control computers are counted, the flight control computers with most tickets exceeding 2/3 are normal, the flight control computers with less tickets than 1/3 are abnormal, and the final output result is the resolving result of the flight control computer with the largest ticket number.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.