VTOL Flight Trajectory Planning With Confidence Intervals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion planning methods for VTOL aircraft, especially in urban environments, face challenges with real-time executability, deterministic decision-making, and high safety requirements, while also incurring high computational and memory costs, making them unsuitable for autonomous civil aviation navigation.
Innovation Solution
A method and system that involves preliminary planning of reference trajectories and emergency trajectories, with confidence intervals defined to allow the aircraft to maneuver within specified spatial volumes, enabling real-time motion planning and actuation of flight control systems based on sensor data and environmental conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complete mission preliminary planning is performed to predict and consider improbable events before they occur, then operational safety is improved, but computational and memory expenditure becomes unmanageable
Solution Approach 1:
The motion planning is divided into two distinct phases: preliminary planning phase (performed before flight) and real-time execution phase (performed during flight). The preliminary planning phase generates a reference trajectory and identifies key waypoints, while the real-time phase only performs local trajectory adjustments based on current sensor data. This segmentation reduces computational burden by separating complex global planning from simple local adjustments.
Solution Approach 2:
All complex computational work for trajectory generation, obstacle prediction, and path optimization is performed in advance during the preliminary planning phase. The reference trajectory and associated data structures are computed before flight, allowing the real-time system to operate with minimal computational resources by only executing local adjustments based on pre-computed information.
2Adaptability or versatility
If real-time planning algorithms are executed on the flight platform to provide flexibility, then adaptability to unpredictable environments is improved, but transparency for certification examination deteriorates
Solution Approach 1:
The system separates deterministic preliminary planning (which provides certification transparency) from flexible real-time execution (which provides adaptability). The preliminary planning phase produces a reference trajectory that can be examined and certified, while the real-time phase executes local adjustments that adapt to unpredictable conditions without compromising the certifiable core planning logic.
Solution Approach 2:
The reference trajectory serves as an intermediary between the certifiable preliminary planning and the flexible real-time execution. It provides a deterministic foundation that can be examined by certification authorities while enabling real-time adaptations through local trajectory adjustments based on sensor feedback.
3Adaptability or versatility
If optimization-based or sampling-based planning algorithms are used for real-time planning, then flexibility in handling unpredictable events is improved, but deterministic behavior required for certification deteriorates
Solution Approach 1:
The system separates deterministic trajectory generation (reference trajectory in preliminary planning) from flexible local adjustments (real-time execution). The reference trajectory provides deterministic behavior for certification, while real-time local adjustments handle unpredictable events without requiring complex optimization or sampling algorithms during flight.
Solution Approach 2:
Instead of performing complete optimization-based planning in real-time, the system performs only partial planning - specifically, local trajectory adjustments around pre-computed waypoints. This partial action provides sufficient adaptability for handling unpredictable events while maintaining deterministic behavior through the use of pre-computed reference trajectories.
Data Source
AI summary
A motion planning method and system for aircraft, in particular for separately electrically driven, load-carrying and/or people-carrying multicopters, includes: a motion preliminary planning unit that executes a preliminary planning algorithm using a computer on the ground or on board an aircraft in question, by which algorithm a reference trajectory, emergency trajectories are determined at intervals along the reference trajectory and confidence intervals are determined along the reference trajectory, which confidence intervals specify a spatial volume in which to maneuver without a pre-planned path but that the aircraft can't leave or is able to leave only at predefined locations; a data store in which parameters of the reference trajectory, parameters of the confidence intervals and parameters regarding a permissible deviation of the aircraft from the reference trajectory are stored on the aircraft according to instructions of the motion preliminary planning unit; a real-time control unit on the aircraft for the real-time motion planning using the parameters and the confidence intervals and using system data of the aircraft recorded by sensors; and an actuation apparatus that actuates flight control/regulation apparatuses of the aircraft using control/regulation signals derived from the real-time motion planning, wherein the flight control/regulation apparatuses of the aircraft are actuatable such that the aircraft is moved or is able to be moved either a) along the reference trajectory or b) along a real-time trajectory within a confidence interval or along an emergency trajectory with a given or required deviation from the reference trajectory.