AI summary
Problem
Current biomimetic drones lack flexible attitude control and adaptive landing capabilities on complex surfaces, while traditional model predictive control is too computationally heavy for real-time deployment on lightweight, low-power hardware.
Approach
The authors designed a spherical-shell drone that mimics bird posture and landing, paired with a bio-inspired MPC controller optimized via sparse matrices and multi-path primal-dual neural networks to run efficiently on embedded systems.
Key results
- Spherical-shell design enabling adaptive landing and immediate re-takeoff on complex surfaces
- Bio-inspired parameter pairs generating realistic bird-like attitude trajectories across flight phases
- Sparse matrix and multi-path PDNN optimization reducing RAM usage by 39.3%
- Attitude tracking mean-square error reduced to 0.0042 rad in simulations and real-world tests
Why it matters
Advances lightweight, agile UAV design for complex terrain navigation and real-time embedded biomimetic control.
Abstract
In nature, birds exhibit outstanding attitude control, enabling flexible and efficient takeoff, hovering and landing — capabilities that have not been fully replicated. Thus, we introduce the lightweight bio-inspired rotary-wing drone (L- BIRD). It incorporates a spherical structure, which can imitate birds’ attitude variation and land on complex surfaces adaptively. L-BIRD employs a model predictive control (MPC) framework to enable real-time tracking of bird-like attitude trajectories derived from bio-inspired parameter pairs. To facilitate lightweight deployment on resource-constrained hardware platforms, we improve MPC framework by multi-path primal-dual neural network (PDNN), matrix sparsity and multiplicative optimiza- tion. Experimental results, both in simulations and real-world deployments, demonstrate that L-BIRD realizes accurate and efficient biomimetic attitude control and diverse environmental adaptability. The attitude trajectory mean-square error (MSE) decreases to 0.0042 rad, random access memory (RAM) usage reduces by 39.3%.