Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Novel Feature Learning-based Bio-inspired Neural Network for Real-time Collision-free Rescue of Multi-Robot Systems

Junfei Li, Simon X. Yang

TL;DR

The proposed FLBBINN aims to reduce the computational complexity of the neural network-based approach and enable the feature learning method to achieve real-time responses to environmental changes.

Abstract

Natural disasters and urban accidents drive the demand for rescue robots to provide safer, faster, and more efficient rescue trajectories. In this paper, a feature learning-based bio-inspired neural network (FLBBINN) is proposed to quickly generate a heuristic rescue path in complex and dynamic environments, as traditional approaches usually cannot provide a satisfactory solution to real-time responses to sudden environmental changes. The neurodynamic model is incorporated into the feature learning method that can use environmental information to improve path planning strategies. Task assignment and collision-free rescue trajectory are generated through robot poses and the dynamic landscape of neural activity. A dual-channel scale filter, a neural activity channel, and a secondary distance fusion are employed to extract and filter feature neurons. After completion of the feature learning process, a neurodynamics-based feature matrix is established to quickly generate the new heuristic rescue paths with parameter-driven topological adaptability. The proposed FLBBINN aims to reduce the computational complexity of the neural network-based approach and enable the feature learning method to achieve real-time responses to environmental changes. Several simulations and experiments have been conducted to evaluate the performance of the proposed FLBBINN. The results show that the proposed FLBBINN would significantly improve the speed, efficiency, and optimality for rescue operations.

A Novel Feature Learning-based Bio-inspired Neural Network for Real-time Collision-free Rescue of Multi-Robot Systems

TL;DR

The proposed FLBBINN aims to reduce the computational complexity of the neural network-based approach and enable the feature learning method to achieve real-time responses to environmental changes.

Abstract

Natural disasters and urban accidents drive the demand for rescue robots to provide safer, faster, and more efficient rescue trajectories. In this paper, a feature learning-based bio-inspired neural network (FLBBINN) is proposed to quickly generate a heuristic rescue path in complex and dynamic environments, as traditional approaches usually cannot provide a satisfactory solution to real-time responses to sudden environmental changes. The neurodynamic model is incorporated into the feature learning method that can use environmental information to improve path planning strategies. Task assignment and collision-free rescue trajectory are generated through robot poses and the dynamic landscape of neural activity. A dual-channel scale filter, a neural activity channel, and a secondary distance fusion are employed to extract and filter feature neurons. After completion of the feature learning process, a neurodynamics-based feature matrix is established to quickly generate the new heuristic rescue paths with parameter-driven topological adaptability. The proposed FLBBINN aims to reduce the computational complexity of the neural network-based approach and enable the feature learning method to achieve real-time responses to environmental changes. Several simulations and experiments have been conducted to evaluate the performance of the proposed FLBBINN. The results show that the proposed FLBBINN would significantly improve the speed, efficiency, and optimality for rescue operations.
Paper Structure (23 sections, 21 equations, 12 figures, 3 tables, 3 algorithms)

This paper contains 23 sections, 21 equations, 12 figures, 3 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Problem Statement
  4. Proposed Approaches
  5. Safety and Intelligent Rescue Path Generation
  6. Task Assignment
  7. Path Planning
  8. Environmental Feature Learning via Neurodynamics
  9. Extraction and Filtering of Feature Neurons
  10. Feature Representativeness and Neurodynamics-based Feature Matrix
  11. Feature Neuron Optimization
  12. Fast Path Planning via Neurodynamics-based Feature Matrix
  13. Simulation Results
  14. Rescue Performance in a Static Environment
  15. Rescue Performance with Moving Obstacles
  16. ...and 8 more sections

Figures (12)

  • Figure 1: Illustration of the robot model.
  • Figure 2: Illustrations of the proposed intelligent multi-robot rescue framework.
  • Figure 3: Examples of the bio-inspired neural network. (a) structure of the neural network featuring exclusively local connections; (b) the dynamic landscape of neural activity.
  • Figure 4: Three typical examples of one robot by choosing different parameters $\sigma$ values.
  • Figure 5: Feature neuron extraction and filtering. (a)a schematic representation of the feature neuron extraction process; (b)the secondary distance fusion.
  • ...and 7 more figures