MMH-Planner: Multi-Mode Hybrid Trajectory Planning Method for UAV Efficient Flight Based on Real-Time Spatial Awareness

TL;DR

A multi-mode hybrid trajectory planning mechanism is proposed, which can enhance the planning efficiency by selecting the optimal planning model for trajectory generation based on prior spatial awareness, which outperforms existing state-of-the-art (SOTA) algorithms across multiple metrics.

Abstract

Motion planning is a critical component of intelligent unmanned systems, enabling their complex autonomous operations. However, current planning algorithms still face limitations in planning efficiency due to inflexible strategies and weak adaptability. To address this, this paper proposes a multi-mode hybrid trajectory planning method for UAVs based on real-time environmental awareness, which dynamically selects the optimal planning model for high-quality trajectory generation in response to environmental changes. First, we introduce a goal-oriented spatial awareness method that rapidly assesses flight safety in the upcoming environments. Second, a multi-mode hybrid trajectory planning mechanism is proposed, which can enhance the planning efficiency by selecting the optimal planning model for trajectory generation based on prior spatial awareness. Finally, we design a lazy replanning strategy that triggers replanning only when necessary to reduce computational resource consumption while maintaining flight quality. To validate the performance of the proposed method, we conducted comprehensive comparative experiments in simulation environments. Results demonstrate that our approach outperforms existing state-of-the-art (SOTA) algorithms across multiple metrics, achieving the best performance particularly in terms of the average number of planning iterations and computational cost per iteration. Furthermore, the effectiveness of our approach is further verified through real-world flight experiments integrated with a self-developed intelligent UAV platform.

Figures (7)

• Figure 1: Framework of the multi-mode hybrid planner.
• Figure 2: Schematic diagram of the goal-oriented spatial awareness method. This figure illustrates three distinct perception scenarios. Specifically, the regions corresponding to green rays indicate no obstacles are detected, representing a safe environment. The areas covered by yellow rays reveal passage through narrow regions via local perception, while red rays denote scenarios where obstacle checking confirms that the rays pass through obstacles. Both the latter two cases involve obstacles that compromise the current flight safety, thereby necessitating obstacle avoidance planning.
• Figure 3: Schematic diagram of the multi-mode hybrid planning mechanism. When no obstacles are detected by the sensing ray, as shown in (a), the system operates in fast planning mode to rapidly generate high-quality trajectories (red curves) without incorporating safety constraints. Conversely, when obstacles are detected (indicated by red collision points in (b)), the system switches to standard optimization mode. In this mode, a spatiotemporally optimized avoidance trajectory is generated within the SFC constructed based on an initial collision-free geometric path. Once newly emerged obstacles (blue solid triangles in (c)) interfere with the ongoing flight trajectory, the emergency obstacle avoidance mode is activated immediately, triggering local re-optimization to promptly compute a safe and dynamically feasible alternative trajectory (green curve).
• Figure 4: Flight trajectory comparison of the four methods in different environments. The red, pink, green, and blue trajectories correspond to the proposed method, EGO-Planner, ROTP, and FastPlanner, respectively. It should be noted that for scenario with 0 obs./m$^2$ (i.e., an obstacle-free environment), the trajectories generated by all methods are straight lines and completely overlap. Therefore, comparative trajectory plots for this scenario are not displayed.
• Figure 5: Performance trends of the four planning methods in different environments: an evaluation of flight time, flight distance, flight energy, and the number of replanning iterations.