Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Hybrid Decision Making for Scalable Multi-Agent Navigation: Integrating Semantic Maps, Discrete Coordination, and Model Predictive Control

Koen de Vos, Elena Torta, Herman Bruyninckx, Cesar Lopez Martinez, Rene van de Molengraft

Abstract

This paper presents a framework for multi-agent navigation in structured but dynamic environments, integrating three key components: a shared semantic map encoding metric and semantic environmental knowledge, a claim policy for coordinating access to areas within the environment, and a Model Predictive Controller for generating motion trajectories that respect environmental and coordination constraints. The main advantages of this approach include: (i) enforcing area occupancy constraints derived from specific task requirements; (ii) enhancing computational scalability by eliminating the need for collision avoidance constraints between robotic agents; and (iii) the ability to anticipate and avoid deadlocks between agents. The paper includes both simulations and physical experiments demonstrating the framework's effectiveness in various representative scenarios.

Hybrid Decision Making for Scalable Multi-Agent Navigation: Integrating Semantic Maps, Discrete Coordination, and Model Predictive Control

Abstract

This paper presents a framework for multi-agent navigation in structured but dynamic environments, integrating three key components: a shared semantic map encoding metric and semantic environmental knowledge, a claim policy for coordinating access to areas within the environment, and a Model Predictive Controller for generating motion trajectories that respect environmental and coordination constraints. The main advantages of this approach include: (i) enforcing area occupancy constraints derived from specific task requirements; (ii) enhancing computational scalability by eliminating the need for collision avoidance constraints between robotic agents; and (iii) the ability to anticipate and avoid deadlocks between agents. The paper includes both simulations and physical experiments demonstrating the framework's effectiveness in various representative scenarios.

Paper Structure

This paper contains 13 sections, 11 equations, 5 figures, 6 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Method
  3. MPC Formulation
  4. Semantic World Model
  5. Coordination Strategy
  6. Controller Configuration
  7. Validation
  8. Implementation
  9. Validation Scenarios
  10. Validation Metrics
  11. Results
  12. Validation on physical system
  13. Conclusions

Figures (5)

  • Figure 1: Motivating examples for the presented approach
  • Figure 2: A simplified depiction of a T-junction environment. Perimeters of areas ($S_i$) are drawn in black dashed lines. Boundaries of areas are drawn in gray. Interfaces ($I_j$) between areas are drawn in red.
  • Figure 3: (a) Simplified structured environment consisting of three areas ($S_0$, $S_1$ and $S_2$), and two sub-areas ($S_{00}$ and $S_{01}$ sub-areas of $S_0$). (b) two different views of the same physical area. (c) Excerpt of a graph world model describing the environment in (a). The colors of the nodes signify their different types.
  • Figure 4: Experimental validation on physical robotic hardware
  • Figure 5: Agents deployed in the validation scenario environments. The grey areas signify drivable space. The colored polygons signify the considered constraints by the respective agents

Theorems & Definitions (1)

  • Definition 1: Semantic Map