Hey guys, in my project I need to run a machine learning model in Arduino Nano 33 BLE and set the BLEBoolCharacteristics to true when the tflOutputTensor is higher than 0.7. Right now my Arduino Nano 33 BLE always having Mbed crash problem when calling this line TfLiteStatus invokeStatus = tflInterpreter->Invoke(); I have tried without BLE function, and the Arduino Nano 33 BLE able to run the model and show the tflOutputTensor on the serial monitor. I not sure why adding BLE function together with running machine learning model can cause Mbed crashing problem to occur. Below is the ino code, mode.h file and Mbed Error Info.
model.h (142.4 KB)
run_model_arduinoide_4.ino (7.5 KB)
t
++ MbedOS Fault Handler ++
FaultType: HardFault
Context:
R0 : 00000096
R1 : 200014A0
R2 : 00000000
R3 : 20001240
R4 : 00000000
R5 : 00078432
R6 : 00000000
R7 : 0007CF7E
R8 : 00000001
R9 : 00000020
R10 : 00078432
R11 : 20001244
R12 : 00000258
SP : 200139B0
LR : 200014A0
PC : 0002B2B0
xPSR : 210F0000
PSP : 20013948
MSP : 2003FFC0
CPUID: 410FC241
HFSR : 40000000
MMFSR: 00000000
BFSR : 00000000
UFSR : 00000100
DFSR : 00000000
AFSR : 00000000
Mode : Thread
Priv : Privileged
Stack: PSP
-- MbedOS Fault Handler --
++ MbedOS Error Info ++
Error Status: 0x80FF013D Code: 317 Module: 255
Error Message: Fault exception
Location: 0x2B2B0
Error Value: 0x2000B54C
Current Thread: main Id: 0x20013B7C Entry: 0x67713 StackSize: 0x8000 StackMem: 0x2000BB58 SP: 0x200139B0
For more info, visit: https://mbed.com/s/error?error=0x80FF013D&tgt=ARDUINO_NANO33BLE
-- MbedOS Error Info --