Momentum-constrained Hybrid Heuristic Trajectory Optimization Framework with Residual-enhanced DRL for Visually Impaired Scenarios

2026-04-16

AI Summary

This work addresses the challenges of multi-objective optimization, limited generalization, and poor decision interpretability in assistive navigation for visually impaired individuals. To this end, the authors propose a synergistic framework integrating Heuristic Trajectory Sampling Clusters (HTSC) with Momentum-constrained Trajectory Optimization (MTO), augmented by residual-enhanced deep reinforcement learning to improve temporal modeling and generalization. A dual-stage cost modeling approach in both Frenet and Cartesian spaces enables interpretable, joint optimization of comfort, safety, and user preferences. Experimental results demonstrate that the proposed method reduces convergence iterations by nearly 50% and significantly enhances trajectory smoothness, robustness, and safety in complex dynamic environments, while simultaneously lowering computational overhead.

Technology Category

• Search and Optimization: Heuristic Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Safe and efficient assistive planning for visually impaired scenarios remains challenging, since existing methods struggle with multi-objective optimization, generalization, and interpretability. In response, this paper proposes a Momentum-Constrained Hybrid Heuristic Trajectory Optimization Framework (MHHTOF). To balance multiple objectives of comfort and safety, the framework designs a Heuristic Trajectory Sampling Cluster (HTSC) with a Momentum-Constrained Trajectory Optimization (MTO), which suppresses abrupt velocity and acceleration changes. In addition, a novel residual-enhanced deep reinforcement learning (DRL) module refines candidate trajectories, advancing temporal modeling and policy generalization. Finally, a dual-stage cost modeling mechanism (DCMM) is introduced to regulate optimization, where costs in the Frenet space ensure consistency, and reward-driven adaptive weights in the Cartesian space integrate user preferences for interpretability and user-centric decision-making. Experimental results show that the proposed framework converges in nearly half the iterations of baselines and achieves lower and more stable costs. In complex dynamic scenarios, MHHTOF further demonstrates stable velocity and acceleration curves with reduced risk, confirming its advantages in robustness, safety, and efficiency.

Problem

• visually impaired assistance
• trajectory optimization
• multi-objective optimization
• generalization
• interpretability

Innovation

• Momentum-Constrained Optimization
• Residual-enhanced DRL
• Hybrid Heuristic Trajectory Planning
• Dual-stage Cost Modeling
• Visually Impaired Assistance