AERO-MPPI: Anchor-Guided Ensemble Trajectory Optimization for Agile Mapless Drone Navigation
Xin Chen, Rui Huang, Longbin Tang, Lin Zhao
AI summary
Problem
Conventional mapping-based drone navigation pipelines incur high computational costs and propagate estimation errors, while single MPPI planners struggle with local minima and poor exploration in dense obstacles.
Approach
The method extracts look-ahead anchor points from a multi-resolution LiDAR representation to generate diverse guiding trajectories, which seed parallel MPPI optimizers running on a GPU to evaluate and select the safest, most efficient path.
Key results
- Achieves real-time onboard planning on an NVIDIA Jetson Orin NX
- Sustains reliable flight above 7 m/s with over 80% success rates in simulation
- Produces smoother trajectories and avoids local traps compared to state-of-the-art baselines
- Validated in real-world experiments across complex, cluttered environments
Why it matters
Enables high-speed, reliable autonomous flight in unknown 3D spaces without heavy mapping overhead, advancing practical deployment of agile aerial robots.
Abstract
Agile mapless navigation in cluttered 3D envi- ronments poses significant challenges for autonomous drones. Conventional mapping–planning–control pipelines incur high computational cost and propagate estimation errors. We present AERO-MPPI, a fully GPU-accelerated framework that unifies perception and planning through an anchor-guided ensemble of Model Predictive Path Integral (MPPI) optimizers. Specifically, we design a multi-resolution LiDAR point-cloud representa- tion that rapidly extracts spatially distributed “anchors” as look-ahead intermediate endpoints, from which we construct polynomial trajectory guides to explore distinct homotopy path classes. At each planning step, we run multiple MPPI instances in parallel and evaluate them with a two-stage multi-objective cost that balances collision avoidance and goal reaching. Imple- mented entirely with NVIDIA Warp GPU kernels, AERO-MPPI achieves real-time onboard operation and mitigates the local- minima failures of single-MPPI approaches. Extensive simula- tions in forests, verticals, and inclines demonstrate sustained reliable flight above 7 m/s, with success rates above 80% and smoother trajectories compared to state-of-the-art baselines. Real-world experiments on a LiDAR-equipped quadrotor with NVIDIA Jetson Orin NX 16G confirm that AERO-MPPI runs in real time onboard and consistently achieves safe, agile, and ro- bust flight in complex cluttered environments. Code is available at https://github.com/XinChen-stars/AERO_MPPI.