Merge branch 'master' into rolling

This commit is contained in:
CaCO3
2022-11-25 16:49:34 +01:00
committed by GitHub
11 changed files with 204 additions and 135 deletions

View File

@@ -62,7 +62,6 @@ xxx @jomjol, TODO: are those URLS correct? The `update.zip` does not contain a `
- n.a.
## [12.0.1](https://github.com/jomjol/AI-on-the-edge-device/releases/tag/v12.0.1), 2022-09-29
Improve **u**ser e**x**perience

View File

@@ -11,8 +11,6 @@
____
#### #33 Implement MATTER protocoll
* see [#1404](https://github.com/jomjol/AI-on-the-edge-device/issues/1404)
@@ -35,6 +33,9 @@ ____
accept the new value and calculate correctly the difference.
(see line 739 onwards in ClassFlowPostProcessing.cpp)
#### ~~#29 Add favicon and use the hostname for the website~~- implemented v11.3.1
~~* https://github.com/jomjol/AI-on-the-edge-device/issues/927~~
#### #28 Improved error handling for ROIs
@@ -70,28 +71,53 @@ ____
#### ~~#22 Direct hint to the different neural network files in the other repositories~~- implemented >v11.3.1
~~* https://github.com/jomjol/AI-on-the-edge-device/issues/644~~
#### #21 Extended "CheckDigitalConsistency" Logik
* https://github.com/jomjol/AI-on-the-edge-device/issues/590
#### #20 Deep sleep and push mode
* Let the device be normally in deep sleep state, and wake it up periodically to collect data and push it via MQTT or HTTP post.
* Support ESP-NOW to reduce the overhead of connecting to wifi and mqtt
* the above should enable battery powered applications
* An other way to set deep sleep would be to enable it in a specific period (at night).
#### #19 Extended log informations
* https://github.com/jomjol/AI-on-the-edge-device/issues/580
#### ~~#18 Document WLAN-strength in web page~~
* ~~https://github.com/jomjol/AI-on-the-edge-device/issues/563~~
#### ~~#17 Direct InfluxDB connection~~
* ~~Done in v10.6.0~~
#### #16 Serial Communication
* https://github.com/jomjol/AI-on-the-edge-device/issues/512
* Send the readout value via RX/TX interface with a dedicated TAG
* Make dedicated communication FlowModule
* Modification of RX/TX communication
* Configuration interfache
#### #15 Calibration for FishEye image
* https://github.com/jomjol/AI-on-the-edge-device/issues/507
@@ -104,6 +130,16 @@ ____
#### ~~#14 Backup and restore option for configuration~~- implemented v11.3.1
* ~~https://github.com/jomjol/AI-on-the-edge-device/issues/459~~
* ~~Implement a zip file compression for store and restore~~
* ~~Update the html to handle it~~
#### #13 Manage non linear gauge without CNN re-training
* https://github.com/jomjol/AI-on-the-edge-device/issues/443
@@ -112,6 +148,12 @@ ____
#### ~~#12 Less reboots due to memory leakage~~
* ~~Issue: #414 & #425 #430~~
#### #11 MQTT - configurable payload
* https://github.com/jomjol/AI-on-the-edge-device/issues/344
@@ -124,14 +166,18 @@ ____
#### #9 Basic auth for the UI
* https://github.com/jomjol/AI-on-the-edge-device/issues/283
* Implementation of an authentication mechanism.
#### #8 MQTT configurable readout intervall
Make the readout intervall configurable via MQTT.
* Change the mqtt part to receive and process input and not only sending
#### #7 Extended Error Handling
Check different types of error (e.g. tflite not availabe) and generate an error on the html page.
@@ -142,6 +188,14 @@ To do:
* Implement a checking algo
* Extend the firmware and html page for the error handling
#### ~~#6 Check for double ROI names~~ - implemented v8.0.0
~~Check during configuration, that ROI names are unique.~~
~~To do:~~
* ~~Implementation of ROI name checking in html code before saving analog or digital ROIs~~
#### #5 Configurable decimal separator (point or comma)
@@ -156,112 +210,6 @@ To do:
#### #2 MQTT-controll with callback
* https://github.com/jomjol/AI-on-the-edge-device/issues/105
Extend the MQTT client to also enable callbacks for configuration setting
To do:
* implement callback for receiving information and override `config.ini` settings
* change configuration management to handle online updates (currently changes need a restart)
* think about the startup, as there the default config is loaded
____
# Refused
#### #9 Basic auth for the UI
* https://github.com/jomjol/AI-on-the-edge-device/issues/283
* Implementation of an authentication mechanism.
This will not be part of the free version
#### #20 Deep sleep and push mode
* Let the device be normally in deep sleep state, and wake it up periodically to collect data and push it via MQTT or HTTP post.
* Support ESP-NOW to reduce the overhead of connecting to wifi and mqtt
* the above should enable battery powered applications
* An other way to set deep sleep would be to enable it in a specific period (at night).
Technically not possible, as the limiting factor for the power consumption is not the esp32, but the electrical curcuit on the ESP32CAM board, which consumes a lot already in deep sleep mode.
# Closed
#### ~~#29 Add favicon and use the hostname for the website~~- implemented v11.3.1
#### ~~#19 Extended log informations~~
~~* https://github.com/jomjol/AI-on-the-edge-device/issues/580~~
#### ~~#29 Add favicon and use the hostname for the website~~- implemented v11.3.1
~~* https://github.com/jomjol/AI-on-the-edge-device/issues/927~~
#### ~~#22 Direct hint to the different neural network files in the other repositories~~- implemented >v11.3.1
~~* https://github.com/jomjol/AI-on-the-edge-device/issues/644~~
#### ~~#18 Document WLAN-strength in web page~~
* ~~https://github.com/jomjol/AI-on-the-edge-device/issues/563~~
#### ~~#17 Direct InfluxDB connection~~
* ~~Done in v10.6.0~~
#### ~~#14 Backup and restore option for configuration~~- implemented v11.3.1
* ~~https://github.com/jomjol/AI-on-the-edge-device/issues/459~~
* ~~Implement a zip file compression for store and restore~~
* ~~Update the html to handle it~~
#### ~~#12 Less reboots due to memory leakage~~
* ~~Issue: #414 & #425 #430~~
#### ~~#6 Check for double ROI names~~ - implemented v8.0.0
~~Check during configuration, that ROI names are unique.~~
~~To do:~~
* ~~Implementation of ROI name checking in html code before saving analog or digital ROIs~~
#### ~~#4 Initial Shifting and Rotation~~ - implemented v7.0.0
* ~~https://github.com/jomjol/AI-on-the-edge-device/issues/123~~
@@ -288,6 +236,27 @@ Technically not possible, as the limiting factor for the power consumption is no
____
#### #2 MQTT-controll with callback
* https://github.com/jomjol/AI-on-the-edge-device/issues/105
Extend the MQTT client to also enable callbacks for configuration setting
To do:
* implement callback for receiving information and override `config.ini` settings
* change configuration management to handle online updates (currently changes need a restart)
* think about the startup, as there the default config is loaded
____
#### ~~#1 Optional GPIO for external flash/lighting~~ - implemented (v8.0.0)
* ~~https://github.com/jomjol/AI-on-the-edge-device/issues/133~~
@@ -301,6 +270,3 @@ Technically not possible, as the limiting factor for the power consumption is no
* ~~Implementation of a software module for external light source (e.g. WS8132 LED controller, ...)~~
* ~~Update of the camera module to use the external light instead of the internal flash light~~
* ~~Adopt the configuration algorithm with a configurable light source~~

View File

@@ -269,8 +269,7 @@ private:
std::unique_ptr< Rgb[] > _secondBuffer;
Rgb *_buffer;
SemaphoreHandle_t _finishedFlag;
// xSemaphoreHandle _finishedFlag;
xSemaphoreHandle _finishedFlag;
int _pixelPosition;
int _componentPosition;

View File

@@ -955,7 +955,7 @@ std::string unzip_new(std::string _in_zip_file, std::string _target_zip, std::st
zw = std::string(archive_filename);
ESP_LOGD(TAG, "Rohfilename: %s", zw.c_str());
if (getFileType(zw) == "BIN")
if (toUpper(zw) == "FIRMWARE.BIN")
{
zw = _target_bin + zw;
ret = zw;

View File

@@ -415,7 +415,7 @@ string ClassFlowControll::GetPrevalue(std::string _number)
std::string ClassFlowControll::UpdatePrevalue(std::string _newvalue, std::string _numbers, bool _extern)
{
double zw;
float zw;
char* p;
_newvalue = trim(_newvalue);
@@ -427,7 +427,7 @@ std::string ClassFlowControll::UpdatePrevalue(std::string _newvalue, std::string
}
else
{
zw = strtod(_newvalue.c_str(), &p);
zw = strtof(_newvalue.c_str(), &p);
if (zw == 0)
return "- Error in String to Value Conversion!!! Must be of format value=123.456";
}

View File

@@ -56,7 +56,6 @@ std::string ClassFlowPostProcessing::GetJSON(std::string _id, std::string _mac,
else
json += " \"value\": \"\"," + _lineend;
json += " \"raw\": \"" + NUMBERS[i]->ReturnRawValue + "\"," + _lineend;
json += " \"pre\": \"" + NUMBERS[i]->ReturnPreValue + "\"," + _lineend;
json += " \"error\": \"" + NUMBERS[i]->ErrorMessageText + "\"," + _lineend;
if (NUMBERS[i]->ReturnRateValue.length() > 0)
json += " \"rate\": " + NUMBERS[i]->ReturnRateValue + "," + _lineend;
@@ -103,7 +102,7 @@ void ClassFlowPostProcessing::SetPreValue(double zw, string _numbers, bool _exte
if (NUMBERS[j]->name == _numbers)
{
NUMBERS[j]->PreValue = zw;
NUMBERS[j]->ReturnPreValue = RundeOutput(zw, NUMBERS[j]->Nachkomma);
NUMBERS[j]->ReturnPreValue = std::to_string(zw);
NUMBERS[j]->PreValueOkay = true;
if (_extern)
{

View File

@@ -1,3 +1,2 @@
manifest_hash: 4e37bb0f9273c4de05f38688720fe32aa6e5b892452694a4f7a2ca1659f02cf6
target: esp32
manifest_hash: 45994dbfed009907994c31f6d279c5861a1eacbf219ce8b58e74e39b3393816a
version: 1.0.0

View File

@@ -121,8 +121,6 @@ bool Init_NVS_SDCard()
void task_NoSDBlink(void *pvParameter)
{
// esp_rom_gpio_pad_select_gpio(BLINK_GPIO);
gpio_pad_select_gpio(BLINK_GPIO);
gpio_set_direction(BLINK_GPIO, GPIO_MODE_OUTPUT);

View File

@@ -13,12 +13,14 @@
src_dir = main
[env:esp32cam]
;platform = espressif32@4.4.0
platform = espressif32@5.2.0
platform = espressif32@4.4.0
;platform = espressif32@5.1.0
;platform = espressif32
board = esp32cam
;board = m5stack-core-esp32
framework = espidf
;board_build.partitions = partitions_singleapp.csv
board_build.partitions = partitions.csv
lib_deps =

View File

@@ -0,0 +1,107 @@
#include <unity.h>
#include <ClassFlowCNNGeneral.h>
class UnderTestCNN : public ClassFlowCNNGeneral {
public:
using ClassFlowCNNGeneral::ZeigerEvalAnalogNeu;
using ClassFlowCNNGeneral::ZeigerEvalHybridNeu;
using ClassFlowCNNGeneral::ClassFlowCNNGeneral;
};
/**
* @brief test if all combinations of digit
* evaluation are running correctly
*/
void test_ZeigerEval()
{
UnderTestCNN undertest = UnderTestCNN(nullptr, Digital100);
// the 5.2 is already above 5.0 and the previous digit too (3)
printf("Test 5.2, 3\n");
int result = undertest.ZeigerEvalAnalogNeu(5.2, 3);
TEST_ASSERT_EQUAL(5, result);
// the 5.2 is already above 5.0 and the previous digit not (9)
// so the current digit shoult be reduced (4.9)
printf("Test 5.2, 9\n");
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalAnalogNeu(5.2, 9));
printf("Test 4.4, 9\n");
// the 4.4 (digital100) is not above 5 and the previous digit (analog) too (9.3)
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalAnalogNeu(4.4, 9));
printf("Test 4.5, 0\n");
// the 4.5 (digital100) is not above 5 and the previous digit (analog) too (9.6)
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalAnalogNeu(4.5, 0));
}
/**
* @brief test if all combinations of digit
* evaluation are running correctly
*/
void test_ZeigerEvalHybrid() {
UnderTestCNN undertest = UnderTestCNN(nullptr, Digital100);
// the 5.2 and no previous should round down
printf("ZeigerEvalHybridNeu(5.2, 0, -1)\n");
TEST_ASSERT_EQUAL(5, undertest.ZeigerEvalHybridNeu(5.2, 0, -1));
// the 5.3 and no previous should trunc to 5
printf("ZeigerEvalHybridNeu(5.3, 0, -1)\n");
TEST_ASSERT_EQUAL(5, undertest.ZeigerEvalHybridNeu(5.3, 0, -1));
printf("ZeigerEvalHybridNeu(5.7, 0, -1)\n");
// the 5.7 and no previous should trunc to 5
TEST_ASSERT_EQUAL(6, undertest.ZeigerEvalHybridNeu(5.7, 0, -1));
// the 5.8 and no previous should round up to 6
printf("ZeigerEvalHybridNeu(5.8, 0, -1)\n");
TEST_ASSERT_EQUAL(6, undertest.ZeigerEvalHybridNeu(5.8, 0, -1));
// the 5.7 with previous and the previous between 0.3-0.5 should round up to 6
TEST_ASSERT_EQUAL(6, undertest.ZeigerEvalHybridNeu(5.7, 0.4, 1));
// the 5.3 with previous and the previous between 0.3-0.7 should round down to 5
TEST_ASSERT_EQUAL(5, undertest.ZeigerEvalHybridNeu(5.3, 0.7, 1));
// the 5.3 with previous and the previous <=0.5 should trunc to 5
TEST_ASSERT_EQUAL(5, undertest.ZeigerEvalHybridNeu(5.3, 0.1, 1));
// the 5.3 with previous and the previous >=9.5 should reduce to 4
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalHybridNeu(5.3, 9.6, 9));
// the 5.7 with previous and the previous >=9.5 should trunc to 5
TEST_ASSERT_EQUAL(5, undertest.ZeigerEvalHybridNeu(5.7, 9.6, 9));
// the 4.5 (digital100) is not above 5 and the previous digit (analog) not over Zero (9.6)
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalHybridNeu(4.5, 9.6, 0));
// the 4.5 (digital100) is not above 5 and the previous digit (analog) not over Zero (9.6)
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalHybridNeu(4.5, 9.6, 9));
// the 4.5 (digital100) is not above 5 and the previous digit (analog) not over Zero (9.5)
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalHybridNeu(4.5, 9.5, 9));
// 59.96889 - Pre: 58.94888
// 8.6 : 9.8 : 6.7
// the 4.4 (digital100) is not above 5 and the previous digit (analog) not over Zero (9.5)
TEST_ASSERT_EQUAL(8, undertest.ZeigerEvalHybridNeu(8.6, 9.8, 9));
// pre = 9.9 (0.0 raw)
// zahl = 1.8
TEST_ASSERT_EQUAL(2, undertest.ZeigerEvalHybridNeu(1.8, 9.0, 9));
// if a digit have an early transition and the pointer is < 9.0
// prev (pointer) = 6.2, but on digital readout = 6.0 (prev is int parameter)
// zahl = 4.6
TEST_ASSERT_EQUAL(4, undertest.ZeigerEvalHybridNeu(4.6, 6.0, 6));
// issue #879 vorgaenger is -1, zahl = 6.7
//TEST_ASSERT_EQUAL(7, undertest.ZeigerEvalHybrid(6.7, -1.0, -1));
}