3D Multicopter Path Following With Independent Yaw Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional pure-pursuit controllers for multicopter drones are limited to 2D path tracking, lack independent yaw control, and struggle with adapting to varying velocities, making them inefficient for 3D applications and high-speed operations.

Innovation Solution

A 3D path follower controller with independent yaw control and adaptive lookahead distance, using heuristic constraints based on multicopter dynamics to regulate velocity commands, enabling precise motion control across a wide velocity range.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If traditional pure-pursuit controllers are used for multicopter drones, then the control system is simple and easy to implement, but the controller is limited to 2D path tracking and cannot operate effectively in 3D space

Engineering Contradiction:
Improvecontroller structureVSAvoidpath tracking dimension
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent extends the traditional 2D pure-pursuit controller to 3D space by introducing a vertical dimension to the path tracking algorithm. The controller now operates in three-dimensional space, allowing multicopter drones to follow complex 3D trajectories including vertical movements, loops, and figure-eight patterns that were impossible with 2D controllers.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Device complexity

If traditional pure-pursuit controllers are used, then the controller design is straightforward, but the controller lacks independent yaw control capability

Engineering Contradiction:
Improvecontroller designVSAvoidyaw control independence
Core Design Contradiction:
Device complexityVSEase of operation

Solution Approach 1:

The patent segments the control system into independent modules: a path follower controller that generates body velocity commands and a separate yaw controller that independently manages heading. This modular segmentation allows each controller to be optimized for its specific function while working together through the cascading architecture, enabling independent yaw control without complicating the overall design.

Inventive Principle:
Principle #1Segmentation

3Device complexity

If traditional pure-pursuit controllers are used, then the controller is easy to implement, but the controller cannot adapt to varying velocities without re-tuning

Engineering Contradiction:
Improvecontroller implementationVSAvoidvelocity adaptation
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic lookahead distance that automatically adjusts based on the drone's current velocity. As the drone speeds up, the lookahead distance increases proportionally, and as it slows down, the distance decreases. This dynamic adaptation eliminates the need for manual re-tuning when operating at different velocities, making the controller inherently adaptive to velocity changes.

Inventive Principle:
Principle #15Dynamics

4Device complexity

If traditional pure-pursuit controllers are used, then the controller structure is simple, but the controller efficiency decreases at high speeds

Engineering Contradiction:
Improvecontroller structureVSAvoidpath following efficiency
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent employs a dynamic lookahead distance that scales with velocity to maintain optimal path following performance across different speed ranges. At higher speeds, the increased lookahead distance allows the controller to anticipate upcoming path changes earlier, improving responsiveness and efficiency without requiring a more complex controller structure.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4575700A1A method, system and computer programs for controlling multicopter drones
Publication Date: 2025.06.25 UM AUTONOMOUS SYSTEMS LTD
  • EP4575700A1 patent drawingFigure 1
  • EP4575700A1 patent drawingFigure 2
  • EP4575700A1 patent drawingFigure 3

AI summary

A method, system and computer programs for controlling multicopter drones are provided. The method comprises performing by one or more processors the following steps: selecting a lookahead point to a multicopter drone on a given 3D path; computing an angular yaw velocity command using a given yaw input; computing linear horizontal and vertical velocity commands given horizontal and vertical distances to the selected lookahead point and a desired cruise velocity; creating an arc going from the multicopter drone to the selected lookahead point by combining the computed angular yaw velocity command with the computed linear horizontal and vertical velocity commands, and computing a velocity vector command using the created arc; and refining the computed velocity vector command based on previous computed velocity commands and an allowed maximum acceleration.