Sequence Modeling for Time-Optimal Quadrotor Trajectory Optimization with Sampling-based Robustness Analysis

2025-06-16

AI Summary

Time-optimal trajectory generation for quadrotors involves computationally expensive non-convex optimization, hindering real-time deployment. Method: This paper proposes an end-to-end trajectory generation framework based on sequential learning: given a collision-free geometric path, it directly predicts the time-optimal velocity profile satisfying dynamic constraints and safety boundaries. A novel backward reachable tube analysis framework guides the model to learn local analytical optimality; input-path random perturbation serves as data augmentation to improve robustness; and a length-agnostic sequential architecture (LSTM/Transformer) enables generalization to arbitrary path lengths. Results: Evaluated on a real quadrotor platform, the method accelerates trajectory generation by over one order of magnitude compared to conventional nonlinear programming solvers, achieves stable millisecond-level inference latency, and maintains high success rates and dynamical feasibility.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Time-optimal trajectories drive quadrotors to their dynamic limits, but computing such trajectories involves solving non-convex problems via iterative nonlinear optimization, making them prohibitively costly for real-time applications. In this work, we investigate learning-based models that imitate a model-based time-optimal trajectory planner to accelerate trajectory generation. Given a dataset of collision-free geometric paths, we show that modeling architectures can effectively learn the patterns underlying time-optimal trajectories. We introduce a quantitative framework to analyze local analytic properties of the learned models, and link them to the Backward Reachable Tube of the geometric tracking controller. To enhance robustness, we propose a data augmentation scheme that applies random perturbations to the input paths. Compared to classical planners, our method achieves substantial speedups, and we validate its real-time feasibility on a hardware quadrotor platform. Experiments demonstrate that the learned models generalize to previously unseen path lengths. The code for our approach can be found here: https://github.com/maokat12/lbTOPPQuad

Problem

• Accelerate time-optimal quadrotor trajectory generation using learning-based models
• Analyze learned models' local properties linked to controller reachability
• Enhance robustness via data augmentation with perturbed input paths

Innovation

• Learning-based models imitate time-optimal planner
• Quantitative framework analyzes learned model properties
• Data augmentation enhances robustness with perturbations