我想使用STM32F0和Truestudio解析NMEA语句中的GNSS数据。

Question

我在从NMEA语句中解析GPS数据时遇到了困难。我正在使用Quectel L89全球导航卫星系统模块(波特率: 115200)和Truestudio IDE。我用CubeMX制作了程序的骨架结构。

我试着将HAL_UART_RxCpltCallback与HAL_UART_Receive_IT和HAL_UART_Receive_IT分开使用。我有时能读数据，有时看不懂。问题是原始数据从$GPRMC开始，以$PSTMCPU结尾，而从ST读取数据时，rec_buffer从$GPRMC(ok条件)开始，有时从$GNVTG(错误条件)开始。

如何确保rec_buffer总是从$GPRMC开始

我在此宣布：(1)

  HAL_UART_Receive_IT(&huart2, rec_buff, 2400);  // from GPS UART

我也试过这个：

void HAL_UART_RxCpltCallback(   UART_HandleTypeDef *    huart   )
{
    if (huart->Instance == USART2)  //GPS UART
    {
        HAL_UART_Transmit(&huart1, &rec_buff[0], 1, 10);
        HAL_UART_Receive_IT(&huart2, rrc_buff, 2400);   
    }
}

L89原始数据：

$GPRMC,094640.000,A,2838.86700,N,07711.56483,E,0.4,99.3,050119,,,A*5E
$GPGGA,094640.000,2838.86700,N,07711.56483,E,1,08,1.0,226.91,M,-35.9,M,,*42
$GNGNS,094640.000,2838.86700,N,07711.56483,E,ANNNNN,08,1.0,0226.9,-35.9,,*78
$GPVTG,99.3,T,,M,0.4,N,0.7,K,A*3D
$GPGST,094640.000,46.0,26.0,17.7,0.2,25.7,18.2,19.6*68
$GPGBS,094640.000,25.7,18.2,19.6,,,,*4B
$GNGSA,A,3,23,09,03,16,26,22,27,07,,,,,1.9,1.0,1.6*28
$GNGSA,A,3,,,,,,,,,,,,,1.9,1.0,1.6*22
$GPGSV,3,1,11,16,70,076,27,03,58,227,30,23,52,331,36,26,44,046,18*78
$GPGSV,3,2,11,22,44,198,21,09,24,316,34,27,19,137,19,07,16,266,27*70
$GPGSV,3,3,11,31,14,058,,14,10,112,,08,09,164,20,,,,*45
$PSTMPRES,14.7,-4.3,3.4,4.7,8.1,-34.9,-10.0,14.5,-8.6,,,,,,,,,,,,,,,,*05
$PSTMVRES,0.3,-0.0,0.0,-0.0,0.3,0.0,-0.5,-0.5,0.1,,,,,,,,,,,,,,,,*0B
$PSTMTG,2034,553618.0000,8,128199118,10,-46760.0000,002a,1025,0,0,10,2034,553618.0000,10,2034,553618.0000,6*5A
$PSTMTS,1,23,87178102.875,-44634.02,01,36,107871,1,9650745.91,13059080.41,21268551.81,-1601.01,2104.54,-615.14,-60384.66,6.69,0,0.00,0.00,0,0,0,0*01
$PSTMTS,1,09,88801609.812,-43889.80,01,34,102805,1,16028885.38,3193585.53,20890300.53,-1492.86,2210.83,804.75,142511.90,11.66,0,0.00,0.00,0,0,0,0*07
$PSTMTS,1,03,86575601.125,-48258.84,01,30,78613,1,13348561.56,22507400.16,4720188.47,-3.99,659.00,-3098.78,52480.77,6.38,0,0.00,0.00,0,0,0,0*2D
$PSTMTS,1,16,85874269.688,-47283.16,01,27,75521,1,-1886095.16,22466622.38,13519219.09,-747.65,-1602.10,2568.93,-3251.94,5.72,0,0.00,0.00,0,0,0,0*2A
$PSTMTS,1,26,3381954.688,-49146.08,00,18,53036,1,-8037289.25,15873771.62,19660297.47,-1174.67,-2217.50,1326.00,26300.98,7.53,0,0.00,0.00,0,0,0,0*3C
$PSTMTS,1,22,3609068.188,-49215.92,00,21,27316,1,10235851.56,24348322.31,-2780494.62,-516.89,-162.93,-3068.35,-184537.24,7.77,0,0.00,0.00,0,0,0,0*08
$PSTMTS,1,27,5418358.750,-44043.68,00,19,77495,1,-10438473.00,23018353.00,-7765070.44,-768.74,651.32,3023.75,-13120.58,14.66,0,0.00,0.00,0,0,0,0*23
$PSTMTS,1,07,90050798.438,-44830.90,01,27,76584,1,25565687.72,6701073.97,4735275.25,-590.77,171.17,3061.98,17221.07,16.12,0,0.00,0.00,0,0,0,0*04
$PSTMTS,1,08,454836.875,-43869.36,00,20,3383,0,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0,0.00,0.00,0,0,0,0*2B
$PSTMNOTCHSTATUS,3786017,0,2449,0,2,5420680,0,3824,0,2*57
$PSTMADCDATA,943,903,,,,,,*48
$PSTMANTENNASTATUS,0*4D
$PSTMSBAS,0,0,,,,*19
$PSTMCPU,48.64,-1,49*41

如有任何帮助/建议，将不胜感激。

谢谢clifford的建议。我需要解析来自RMC、VTG、GGA和GSA的数据。我不能改变波特率，也不能停止专有的句子。

Answer 1

您的GNSS输出没有什么“错误”；这种设备输出许多标准和专有语句是正常的。您的职责是要么将模块配置为只输出所需的句子，要么解析这些句子并丢弃那些您不想要的句子。

向GNSS发送以下命令以禁用除RMC语句之外的所有可配置语句：

$PSMTSETPAR,1201,0X00000040*75

您需要在每次启动时发送此配置，或者通过以下方式将配置保存在非易失性内存中：

$PSMTSAVEPAR*58

该模块可能会输出其他不可配置的句子(我不是特定模块的familiare )；您可能仍然需要在主机上执行解析和筛选。这就像解析说话者ID和格式说明符以及丢弃你不感兴趣的句子一样简单。有很多方法可以做到这一点；你的问题没有显示出这样的尝试。

您的代码太“低级”了- IEC61162-1 (NMEA 0183)语句格式是面向行的，但是HAL_UART_Receive_IT()只会在不了解协议的情况下获取一大块数据，并且与句子结构完全异步--您可以很容易地开始和结束中段语句。此外，HAL_UART_RxCpltCallback()发生在中断上下文中，而不是接收大量数据并对其进行解析的地方。您首先需要一些更高级的串行I/O代码，它可以将传入的数据缓冲到队列中，并支持面向行的输入。

然后，您需要处理每一行--丢弃那些没有兴趣的行。从这个意义上说，这是一个过于宽泛的问题，与GNSS数据解析无关--这实际上不是您的问题，正如我所说的，它的级别低于-您需要构建一个合适的串行I/O基础结构。

另一个问题是，等待接收2400个字符，您将得到块的数据，并且只有最近收到的数据才是有效的。如果需要实时接收当前职位，则不能这样做。

我不是圣HAL的专家，不会选择使用它；它是很糟糕的记录。在HAL_UART_RxCpltCallback中，您应该做的是将接收到的数据放入FIFO缓冲区中，由其他线程处理(即不在中断上下文中处理)；由于NMEA语句的长度可变，您需要在单个字符的基础上这样做。一种可能更简单的方法是轮询霍尔接收缓冲区，使其具有零超时，并累积字符直到一行可用为止。然后，您可以检查这一行，看看它是否是RMC语句，然后处理它。例如，以下允许对串行输入进行非阻塞处理.如果不需要执行其他函数，则可以延长超时以无限期地等待每个字符：

// Strictly NMEA0183 requires only 81 byte sentence buffer, but 
// L89 appears to output non-compliant proprietary sentences.
static char line_buffer[1024];
int line_buffer_index = 0;

for (;;)
{
    HAL_StatusTypeDef rx_status = HAL_OK ;
    while( rx_status == HAL_OK )
    {
        // Receive a character if available (zero timeout)
        char ch = 0;
        HAL_StatusTypeDef rx_status = HAL_UART_Receive( &huart2, ch, 1, 0 ) ;

        // If character received and not the LF following the previous CR
        if( rx_status == HAL_OK && ch != `\n ` )
        {
            // If the end of the sentence...
            if( ch == `\r ` )
            {
                // terminate the line
                line_buffer[line_buffer_index] = `\0` ;

                // Start a new line
                line_buffer_index = 0 ;

                // Check if line is an RMC from a GNSS
                if( line_buffer_index > 6 &&
                    memcmp( line_buffer, "$G", 2) == 0 &&
                    memcmp( &line_buffer[3], "RMC", 3) == 0 )
                {
                    // Process RMC
                    processRMC( line_buffer ) ;
                }
            }
            else
            {
                // add character to line buffer
                line_buffer[line_buffer_index] = ch ;
                line_buffer_index++ ;
                if( line_buffer_index > sizeof(line_buffer) - 1 )
                {
                    // Line too long, discard all
                    line_buffer_index = 0 ;
                } 
            }
        }
    }

    // do other work here if necessary
}