MSSP : A Versatile Multi-Scenario Adaptable Intelligent Robot Simulation Platform Based on LIDAR-Inertial Fusion

TL;DR

The paper introduces MSSP, a Gazebo-based, multi-scenario robot simulation platform for LIDAR-inertial fusion that delivers absolute ground-truth positioning and supports both manual and autonomous robot control. It enables comprehensive evaluation of SLAM algorithms across a spectrum of environments and sensor types, including mechanical and solid-state LIDARs, by providing customizable worlds, realistic sensor noise, and ground-truth data within ROS. Through extensive experiments, the authors demonstrate the platform's ability to facilitate detailed algorithm analysis, highlight strengths of algorithms like Fast-LIO2, Faster-LIO, and Voxel-Map, and reveal challenges posed by dynamic objects and limited FoV in solid-state LIDARs. The work emphasizes the platform’s extensibility in sensors, environments, and mapping evaluation, and releases the tool open-source on GitHub, enabling broad adoption and future enhancements such as GPU-accelerated solid-state LIDAR simulation.

Abstract

This letter presents a multi-scenario adaptable intelligent robot simulation platform based on LIDAR-inertial fusion, with three main features: (1 The platform includes an versatile robot model that can be freely controlled through manual control or autonomous tracking. This model is equipped with various types of LIDAR and Inertial Measurement Unit (IMU), providing ground truth information with absolute accuracy. (2 The platform provides a collection of simulation environments with diverse characteristic information and supports developers in customizing and modifying environments according to their needs. (3 The platform supports evaluation of localization performance for SLAM frameworks. Ground truth with absolute accuracy eliminates the inherent errors of global positioning sensors present in real experiments, facilitating detailed analysis and evaluation of the algorithms. By utilizing the simulation platform, developers can overcome the limitations of real environments and datasets, enabling fine-grained analysis and evaluation of mainstream SLAM algorithms in various environments. Experiments conducted in different environments and with different LIDARs demonstrate the wide applicability and practicality of our simulation platform. The implementation of the simulation platform is open-sourced on Github.

Figures (9)

• Figure 1: System architecture of MSSP simulation platform. This platform comprises two core components : the robot system simulation(shown in left pink rectangular box) and the SLAM algorithm evaluation(shown in right yellow rectangular box)
• Figure 2: System architecture of the robot simulation model. The left image shows the overall structure and detailed dimensions of the robot model, while the right one presents detailed information on the sensor suite and motion chassis.
• Figure 3: Bspline Path Generation and Pure Pursuit Algorithm for Robot Autonomous Tracking
• Figure 4: Specific Schematic Diagrams of 10 Simulated Environments, from top left to bottom right, respectively corresponding to museum, factory, hospital, neighborhood, courtyard, corridor, district, warehouse, farmland, and desert scene.
• Figure 5: Map constructed by Faster-LIO algorithm and "ghost" effects caused by moving personnel in warehouse sequence