Arduino-Based Quadcopter Flight Controller: 170-Line Code Explained

Beyond basic tuning, advanced techniques include: (1) Interactive fine-tuning using sliders to adjust transient response aggressiveness and frequency domain parameters (bandwidth, phase margin), (2) Gain scheduling for highly nonlinear systems where dynamics change significantly across operating points (e.g., engine speeds from 2000 to 3000 RPM), involving linearization at multiple operating points and implementing lookup tables for scheduled gains.

Manual tuning techniques for model-based PID include pole placement (placing closed-loop poles at desired locations), loop shaping using Bode/Nyquist plots to shape frequency response, and heuristic methods applied to simulation. Automatic tuning algorithms implement established heuristics within software, generating initial gain sets that typically require manual refinement. Auto-tuning can operate on models, combine with system identification using input-output data, or run in real-time on hardware. Despite automation capabilities, manual tweaking remains essential for achieving precise desired responses. This flexibility makes PID particularly valuable for well-behaved systems where users can later adjust gains for specific situations without requiring extensive control theory knowledge.

PID tuning involves adjusting three parameters—Proportional (P), Integral (I), and Derivative (D)—to achieve stable quadcopter flight: P controls immediate response to errors (too low causes rotation, too high causes violent vibrations); I compensates for accumulated errors like wind (too low causes drift, too high causes constant vibration); D predicts and dampens future movements (too low causes oscillation, too high causes lag). The tuning process requires starting with P at default values, I and D at zero, then systematically adjusting each parameter while making lateral movements to observe the quadcopter's behavior, keeping the battery connected to prevent resetting values.

Effective quadcopter PID tuning follows a systematic methodology: first ensure mechanical soundness of the frame, then fly test flights on default settings to analyze frame resonance using Blackbox logs; configure filters (dynamic notch at 250/100/400 Hz and static bi-quad at 100 Hz), disable gyro low-pass filters, and set pidsum_limit to 1000%; finally, tune D-gain first by incrementally increasing the master multiplier (0.1-0.2 steps) until signs of oscillation or motor strain appear, then back off slightly, as higher D-gains provide better flight response but increase motor heat and flyaway risk.

This comprehensive section covers advanced tuning methodologies and underlying principles. Core topics include: (1) RPM filtering configuration requirements including ESC firmware versions and protocol settings; (2) Gyro RPM filter versus dynamic notch filter functionality and interaction; (3) Recommended RPM filter parameter values; (4) I-term function and its trade-offs between stability and stiffness; (5) PID slider proportional balancing mechanics; (6) Master multiplier technique for adjusting I-term; (7) Stick response gain and flight feel customization; (8) Anti-gravity term function for throttle transition smoothing. The overarching philosophy emphasizes that there is no objectively correct tuning configuration, and pilots should develop their own style based on personal flying preferences and aircraft characteristics.

Fail Safe is a critical safety mechanism that activates when connection is lost between radio and drone. There are two types of connection failures: radio-to-receiver and receiver-to-flight controller. Configure Fail Safe by setting the drone to the desired state (throttle at 0, stabilization mode, motors off), then record this state in the receiver using its button. Telemetry enables the flight controller to send sensor data back to the radio. In Betaflight, configure UART ports for devices like receivers and video transmitters. Use SmartPort protocol for modern devices. Connect receiver telemetry to UART 3 and VTX to UART 6. Enable telemetry output in Betaflight configuration. Discover sensors by selecting 'Discover Sensors' in the radio's telemetry page. Key sensors include RSSI (signal strength) and VFACE (battery voltage).

To create a telemetry control for detecting signal loss, navigate to advanced properties and telemetry controls. Label the control (e.g., 'q') and enable it. Set the type to a switch with the condition 'x equals zero' to activate when telemetry drops to zero. Set a duration of half a second to prevent triggering during momentary dropouts. Set the default position to 100% (on) so that when telemetry is lost, the switch automatically activates. When the battery is disconnected, telemetry drops out and the switch takes its default position (on). When the battery is reconnected, telemetry returns and the switch goes back to off because the condition 'q equals zero' is no longer met.

Fail safe systems are critical safety mechanisms that prevent drone disasters during flight. The two primary causes of control loss are battery depletion and loss of signal between the remote controller and drone. Battery fail safe configuration involves setting a voltage threshold (37-38V for 6-cell batteries) that triggers emergency actions. Voltages below 35V can permanently damage batteries. The configuration uses voltage rather than percentage because voltage data is more commonly available in telemetry systems. When the threshold is reached, the drone emits warning sounds and executes the configured action.

Critical safety systems include: (1) Battery monitoring with analog voltage or current sensing to trigger automatic return-to-home when low, (2) RTK GPS corrections using internet-connected base station networks (like Iowa's Leica Network) for centimeter-level accuracy, (3) Home point establishment for automatic landing location, (4) Fail-safe behavior configuration that returns the drone to home point when battery depletes or other issues occur.

Telemetry notifications provide audio alerts for critical flight conditions including low battery, weak signal, and other parameters. OpenTX displays real-time telemetry data (voltage, current, RSSI) in model settings. Logical switches enable creating conditions that trigger actions when telemetry values meet criteria. Special functions define actions including audio messages, channel changes, timer starts, and light control. Battery monitoring can be configured by voltage thresholds or capacity consumption. Capacity-based alerts track irreversible battery usage. Delay settings prevent false alerts during rapidly changing conditions. Multiple switches can be combined for complex conditions.

Real-Time Operating Systems (RTOS) enable microcontrollers to execute multiple tasks concurrently through time-slicing. The ESP32 Arduino implementation includes an RTOS that allows creating multiple tasks. Tasks are created by defining functions with void return type and void parameters, containing infinite loops (while(1) or for(;;)). Unlike traditional Arduino loops, these functions are managed by the RTOS scheduler, enabling parallel execution of multiple independent code paths.

Porting FreeRTOS to STM32 microcontrollers requires understanding multitasking principles, CPU resource allocation, and memory requirements for RAM and Flash storage to ensure efficient real-time task scheduling and system stability.

This video provides a comprehensive overview of 35 real-time operating systems (RTOS) available for STM32 microcontrollers, with 14 being free or conditionally free. The presenter offers a personal ranking based on practical experience: FreeRTOS (market leader with broad licensing for critical applications like medical and aerospace), MicroC/OS (comprehensive automation features), mbed OS (modular IoT-focused design), and Zephyr (multi-protocol support). Key considerations for selection include licensing terms, debugging capabilities, documentation quality, hardware support breadth, and specific project requirements. The video emphasizes that while many RTOS options exist, the choice depends on factors such as development tools, community support, and the complexity of the target application.

Real-Time Operating Systems (RTOS) are designed to process data within strict time constraints. They are divided into two types: Hard Real-Time Systems (where missing deadlines can cause catastrophic failures, like airbag deployment in cars) and Soft Real-Time Systems (where missing deadlines may cause inconvenience but not catastrophic failures, like camera photography). RTOS requires well-defined fixed time constraints and is used in applications like flight control systems and missile guidance.

A Real-Time Operating System (RTOS) is a lightweight operating system designed for microcontrollers that enables concurrent execution of multiple tasks while guaranteeing strict timing deadlines, unlike general-purpose operating systems which prioritize human responsiveness and may have non-deterministic scheduling; RTOS is particularly useful when developers need to run several independent threads simultaneously on a processor, such as handling user input, network communication, sensor reading, and display updates concurrently on devices like the ESP32, though simpler projects with minimal concurrency needs may be better suited for bare-metal super loop architectures.