Aircraft Vertical Trajectory Optimization Under Real-Time Energy Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating optimized aircraft trajectories in dynamic environments with obstacles require lengthy computation times, leading to increased workload for pilots and potential overshooting of target points during descent and approach phases.
Innovation Solution
A method using a deterministic neural network to rapidly compute an estimated overall cost for each state, allowing for real-time generation of an optimized vertical trajectory by iteratively determining next states and selecting those with the lowest cost, while integrating constraints and atmospheric conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If iterative processing is performed to generate an optimized vertical trajectory considering energy constraints and obstacles, then trajectory optimization quality is improved, but computation time increases
Solution Approach 1:
The patent pre-calculates and stores performance data (fuel consumption, time, distance) for various flight conditions in lookup tables before actual trajectory optimization. During iterative processing, the system retrieves pre-computed values instead of calculating them in real-time, significantly reducing computation time while maintaining optimization quality
Solution Approach 2:
The patent creates simplified models and lookup tables that copy essential performance characteristics of the aircraft without requiring full complex simulations. These simplified representations allow rapid evaluation of trajectory options during iterative optimization, balancing accuracy with computational efficiency
2Productivity
If real-time trajectory generation is implemented using onboard means, then pilot workload is reduced and response time is improved, but computational complexity increases
Solution Approach 1:
The patent divides the trajectory optimization problem into discrete state nodes and transition segments. Each segment represents a specific maneuver or flight phase with pre-characterized performance properties. This segmentation allows the complex optimization problem to be solved through systematic evaluation of manageable segments rather than continuous complex calculations
Solution Approach 2:
The patent introduces lookup tables and pre-computed performance data as intermediaries between the optimization algorithm and actual flight calculations. These intermediaries store pre-analyzed performance characteristics, allowing the system to query results rapidly without performing full complex simulations during real-time operation
Data Source
AI summary
A method and device for generating an optimum aircraft vertical trajectory, including a unit performing iterative processing to determine, on each iteration, a next state from a computational state, by using an estimated overall cost for next states, each estimated overall cost, which is computed by a cost computation unit, being equal to the sum of a real cost computed up to the next state under consideration by using predetermined constraints and a cost estimated up to the current state of the aircraft. The estimated cost is computed using a deterministic neural network based on performance calculations for the aircraft without using energy constraints, allowing computation of this estimated cost and the estimated overall cost to be performed rapidly. The iterative processing is repeated until the determined state is situated in proximity to the current state of the aircraft, the corresponding trajectory part forming the optimum vertical trajectory.