NAVIPAC
RAW RECORDING AND REPLAY
Last update: 02/04/2020
Version: 4.2
Contents
1 General overview ........................................................................................................ 3
2 Raw recording............................................................................................................. 4
2.1 Setting up raw recording ......................................................................................... 4
2.2 Recording raw data ................................................................................................ 5
3 Raw replay .................................................................................................................. 7
4 Raw data format........................................................................................................ 10
Page 2 of 11
1 General overview
The NaviPac raw recording and replay function is a built-in part of NaviPac and coexists
with the ordinary data logging.
Recording
Figure 1 Recording data flow
Replay
Figure 2 Replay data flow
NaviPac_Raw_Recording_And_Replay.docx
Page 3 of 11
Last update: 2/4/2020
As illustrated above, the remaining parts of NaviPac will not see the presence of recording
and replay – to them, it just looks like an ordinary operation.
The raw recording is a power full tool for trouble shooting, so it could a case where the EIVA
help desk ask you to perform a recording
Please note that the raw recording to generate rather huge data files – so consider carefully
when using
2 Raw recording
2.1 Setting up raw recording
The use of NaviPac raw recording must be defined as a part of the general parameter
configuration in NaviPac Project Settings.... Enable this function by checking the checkbox
under Project Settings > Advanced > Special raw logging.
Figure 3 Enabling and setting up raw recording
Page 4 of 11
• Auto Stop/Start
Switches automatically to a new logfile if changes have been made in online setup
(shift priority, weighting, C-O etc.) Recommended: Enabled.
• Special raw logging
Enable the special raw recording function in NaviPac and store data in a special raw
format for possible playback. The function must be started from Online > Options >
NaviPac Raw log menu. Note that the logging requires substantial disc space.
• Auto start raw log module
Include the raw recording module in the Warm start list of which modules to open
automatically upon navigation start, e.g. C:\EIVA\NaviPac\bin\NPRawLog.exe.
If not enabled here, you may start the module manually from Online > Options >
NaviPac Raw Log...
• Auto start raw logging
Start raw recording (as a kind of black box recording) just after start of navigation.
2.2 Recording raw data
Raw recording in NaviPac is performed by the NPRawLog.exe module, which is located in
\eiva\navipac\bin folder. The module can be started from the Online > Options > NaviPac
Raw Log menu item or as a warm start function as described above.
Figure 4 NaviPac Raw Recording module
NaviPac_Raw_Recording_And_Replay.docx
Page 5 of 11
Last update: 2/4/2020
The module contains all required information and parameters:
• Information from kernel:
• NaviPac status: Read-only. Displays whether the NaviPac online
navigation is running or not.
• Packet information: RDID buffer id, NaviPac timer and packet size
• Number of records (message counts).
• #Instruments: Number of instruments included in the NaviPac setup.
• Record Size: Read-only. Number of bytes recorded per cycle.
• Raw recording enabled: Read-only. Displays that the raw recording
function has been enabled.
• Recording setup:
• Path field: Read-only. The location where the data are to be recorded –
change by pressing the Edit button to the right of this field.
• Manual or From LogData control: If you select From LogData, recording
starts/stops automatically upon signal from the Log Data module. If you
select Manual, recording will be controlled by the SOL, EOL and Pause
buttons in this module. The buttons are available/unavailable corresponding
to actual recording status.
• Max size: Read-only. Maximum file size in megabytes. If the value
displayed is ‘0’, then no maximum file size has been specified – for any
other value, NaviPac changes the file automatically when it has reached the
defined limit. Note that the total size of the recorded data file might be a
little bigger than the defined size. To change the maximum size, please
press the editing button to the right of the Max size field and enter the
desired value in the pop-up box that opens.
• Recording status:
• Current file name field: If LogData logging has been selected, it uses same
name as LogData (extension *.NPR). If manual logging has been selected,
it is named YYYYMMDDHHMMSS.NPR.
• Size: Number of data records in the file – corresponds to the number of
NaviPac cycles performed and size of the file in megabytes.
• Header size
• Internal NaviPac information (just informative for the developer team)
Each raw data recording file is totally self-containing, as it besides the data also includes the
internal configuration files required for a replay:
• GENSETUP.DB
• ONLSETUP.DB
• NAVIPAC.INI
• OBJECTS.TXT
For further details on raw data recording file format, please refer to Section 4 Raw data
format.
Page 6 of 11
3 Raw replay
Raw replay is performed via the special module NPRawPlayBack.exe, which is located in
\eiva\navipac\bin. The module must be started manually via the file browser.
Figure 5 NaviPac PlayBack Control module
The first step in the use of the raw replay is to select a file via the Select File button.
Selecting an acceptable file enables display of file information.
File information:
• Total Size: Read-only. Total size of the file in megabytes.
• Record size: Read-only. Size of one data cycle in bytes.
• #Records: Read-only. Number of records in the file – corresponds to the number of
NaviPac cycles performed.
• Date field: Displays start date of the file.
• Time field: Displays start time of the file.
• Configuration files: Size of the four configuration files.
• Extract: Opens special dialogue box to extract the configuration files from the
recorded file.
NaviPac_Raw_Recording_And_Replay.docx
Page 7 of 11
Last update: 2/4/2020
Figure 6 Extracting NaviPac configuration files
In this dialogue box, press Extract for each file you want to extract. Please note – if you are
unsure of whether a file is already extracted or not – extract it. It is the user’s responsibility
to have compatible setup files prior to the playback.
Once the setup files are successfully extracted (or are already compatible), then you may
access the replay functions.
Playback:
• Synchronise time: Select whether NaviPac should set the PC clock to match the
time that the data was recorded. This must NOT be combined with any ordinary
automatic time synchronisation functions. After replay, the PC clock is set back to
the original time.
Page 8 of 11
Replay control button
Figure 7 NaviPac PlayBack Control replay control buttons
• Start button: Start replay. Opens NaviPac online navigation and starts navigating
using default settings (from restored DB and INI files). Opens output ports, enables
recording, etc as if it was an ordinary online setup. If synchronisation is on, data will
be timestamped as when collected.
• Fast forward button: Replay with four times the original speed. Data recording is
less accurate – but computations are valid.
• Pause button: Pause the replay – keep the NaviPac windows open.
• Stop button: Stop replay and close all related NaviPac windows.
• Extract raw data button: Using this tool, you can extract the recorded raw data into
a listing file including header and timestamped data – very similar to the raw data
part of NaviPac LogData G format.
• Play field: Displays the current replay time position.
NaviPac_Raw_Recording_And_Replay.docx
Page 9 of 11
Last update: 2/4/2020
4 Raw data format
The recorded raw data file format is named NPR (NaviPac raw data) and consists of a file
header and a series of raw data records.
The file header is defined as the following, using C language syntax:
class RawLogDataHeader
{
public:
void readHeader(void);
char title[255]; // File name
double timeGetTimeVal; // Time at SOL timeGetTime/100.0
SYSTEMTIME GetSystemTimeVal; // Time at SOL GetSystemTime
int recordSize; // Size of each logged entry
RawLogDataHeader ()
{timeGetTimeVal = 0;title[0] = 0;};
protected:
char mess[100]; // Small note - test only
char objectText[12*1024]; // Copy of objects.txt
char INIFileText[1024*100]; // Copy of NAVIPAC.INI
char gensetupDB[40*1024]; // Copy of gensetup.DB
char onlsetupDB[20*1024]; // Copy of onlsetup.DB
int objTextLen, INILen, onlLen, GenLen;
// Actual size of supporting files
};
This information is always stored as the first record in the file – it may be read using
standard file operations and the size of method (e.g. Read(lh,sizeof(RawLogDataHeader))).
Each time NaviPac performs a cycle, it generates a new record in the file. A record consists
of two parts: a header and the data. The record header is defined as:
typedef struct _myHeader
{
double start_of_cycle, // Local timestamp for start of cycle
cycle_time, // How long time the processing took
nowT; // Time of logging
SYSTEMTIME nowST; // Start of cycle - via GetSystemTime
} MyHeader;
The record header is needed in order to establish the relationship between kernel time
(starting computation) and data time (sensor timestamps) in both PC time (timeGetTime)
and absolute time
Page 10 of 11
The data itself follows the internal structure between the NaviPac dataIO and kernel
modules (copy of shared memory block).
The data layout is described in dataio-kernel.h, and is in short terms:
/*===================================================================
Data from Data I/O to Kernel through shared memory
====================================================================*/
static char RawDataDBKeyPath[3][20] = {"RawDataDB1","RawDataDB2","RawDataDB3"};
#define MAX_INSTANCES 3 // 1 for reading, 1 for writing and one free
#define MAX_PORTS 128 // Maximum ports included
#define MAX_CYCLE_TIME 5 // RawDataDB contains data for maximum 5 seconds
typedef struct // Header information
{
short Port; // Number of port or simulate id
short PacketLength; // Length of current data type
int StartOffset; // Reference to RawDataDB.Packet
short NoPackets; // Number of packets in current cycle
} RawDataDBHead;
typedef struct // One Packet in RawDataDB
{
double TStamp; // Time stamp giving time a la time_t
TIMEVAL TimeAbs; // Time stamp in absolute time
char Data[1]; // RawDataDBHead.PacketLength bytes
} SensorPacket;
typedef struct // One RawDataDB instance
{
int NoInstruments; // Current number of instruments
int KernPID; // Process ID
int DataioPID;
RawDataDBHead Head[MAX_PORTS]; // Header for each port
char Data[1]; // All data for this cycle
} RawDataDB;
Data from one cycle is kept in a RawDataDB, where Head defines the actual layout
(RawDataDBHead defines port number, packet length and start offset) and Data includes all
the sensor data records (SensorPacket gives timestamp and telegram).
The layout and size are fixed from cycle to cycle, so a lot of the space may be left empty.
NaviPac_Raw_Recording_And_Replay.docx
Page 11 of 11
Last update: 2/4/2020
