A New Robot Navigation Algorithm Based on a Double-Layer Ant Algorithm and Trajectory Optimization

TLDR

The paper proposes DL‑ACO, a double‑layer ant colony optimization algorithm for autonomous robot navigation. DL‑ACO uses a two‑stage process: a parallel elite ant colony optimization generates an initial collision‑free path, followed by a turning‑point optimization and B‑spline smoothing to refine length, smoothness, and safety, and is evaluated against other planners in simulation. Experiments and simulations demonstrate that DL‑ACO consistently produces faster, smoother, and safer collision‑free paths than competing planners in both indoor and outdoor settings.

Abstract

This paper presents an efficient double-layer ant colony optimization algorithm, called DL-ACO, for autonomous robot navigation. This DL-ACO consists of two ant colony algorithms that run independently and successively. First, a parallel elite ant colony optimization method is proposed to generate an initial collision-free path in a complex map, and then, we apply a path improvement algorithm called turning point optimization algorithm, in which the initial path is optimized in terms of length, smoothness, and safety. Besides, a piecewise B-spline path smoother is presented for easier tracking control of the mobile robot. Our method is tested by simulations and compared with other path planning algorithms. The results show that our method can generate better collision-free path efficiently and consistently, which demonstrates the effectiveness of the proposed algorithm. Furthermore, its performance is validated by experiments in indoor and outdoor environments.