Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LiDAR-Inertial SLAM-Based Navigation and Safety-Oriented AI-Driven Control System for Skid-Steer Robots

Mehdi Heydari Shahna, Eemil Haaparanta, Pauli Mustalahti, Jouni Mattila

TL;DR

This work tackles autonomous navigation and safe AI-driven control for heavy skid-steer wheeled mobile robots by coupling LiDAR-inertial SLAM with a robust RAID controller. The system combines LiDAR-inertial SLAM (LIO-SAM) for real-time pose estimation, pure-pursuit path-following, inverse kinematics for wheel commands, and a safety-oriented AI controller using radial basis function networks with barrier-function constraints to bound overshoot and steady-state error. A formal stability analysis based on a logarithmic barrier Lyapunov function proves uniform exponential stability under modeling uncertainties and external disturbances, with an integrated design that maintains safety while achieving exponential convergence. Experimental validation on a 4,836 kg MPD under low-light, soft terrain conditions demonstrates accurate navigation, safe actuator limits, and robust tracking compared to alternative controllers, underscoring practical impact for safe AI-driven control of heavy MRNs.

Abstract

Integrating artificial intelligence (AI) and stochastic technologies into the mobile robot navigation and control (MRNC) framework while adhering to rigorous safety standards presents significant challenges. To address these challenges, this paper proposes a comprehensively integrated MRNC framework for skid-steer wheeled mobile robots (SSWMRs), in which all components are actively engaged in real-time execution. The framework comprises: 1) a LiDAR-inertial simultaneous localization and mapping (SLAM) algorithm for estimating the current pose of the robot within the built map; 2) an effective path-following control system for generating desired linear and angular velocity commands based on the current pose and the desired pose; 3) inverse kinematics for transferring linear and angular velocity commands into left and right side velocity commands; and 4) a robust AI-driven (RAID) control system incorporating a radial basis function network (RBFN) with a new adaptive algorithm to enforce in-wheel actuation systems to track each side motion commands. To further meet safety requirements, the proposed RAID control within the MRNC framework of the SSWMR constrains AI-generated tracking performance within predefined overshoot and steady-state error limits, while ensuring robustness and system stability by compensating for modeling errors, unknown RBF weights, and external forces. Experimental results verify the proposed MRNC framework performance for a 4,836 kg SSWMR operating on soft terrain.

LiDAR-Inertial SLAM-Based Navigation and Safety-Oriented AI-Driven Control System for Skid-Steer Robots

TL;DR

This work tackles autonomous navigation and safe AI-driven control for heavy skid-steer wheeled mobile robots by coupling LiDAR-inertial SLAM with a robust RAID controller. The system combines LiDAR-inertial SLAM (LIO-SAM) for real-time pose estimation, pure-pursuit path-following, inverse kinematics for wheel commands, and a safety-oriented AI controller using radial basis function networks with barrier-function constraints to bound overshoot and steady-state error. A formal stability analysis based on a logarithmic barrier Lyapunov function proves uniform exponential stability under modeling uncertainties and external disturbances, with an integrated design that maintains safety while achieving exponential convergence. Experimental validation on a 4,836 kg MPD under low-light, soft terrain conditions demonstrates accurate navigation, safe actuator limits, and robust tracking compared to alternative controllers, underscoring practical impact for safe AI-driven control of heavy MRNs.

Abstract

Integrating artificial intelligence (AI) and stochastic technologies into the mobile robot navigation and control (MRNC) framework while adhering to rigorous safety standards presents significant challenges. To address these challenges, this paper proposes a comprehensively integrated MRNC framework for skid-steer wheeled mobile robots (SSWMRs), in which all components are actively engaged in real-time execution. The framework comprises: 1) a LiDAR-inertial simultaneous localization and mapping (SLAM) algorithm for estimating the current pose of the robot within the built map; 2) an effective path-following control system for generating desired linear and angular velocity commands based on the current pose and the desired pose; 3) inverse kinematics for transferring linear and angular velocity commands into left and right side velocity commands; and 4) a robust AI-driven (RAID) control system incorporating a radial basis function network (RBFN) with a new adaptive algorithm to enforce in-wheel actuation systems to track each side motion commands. To further meet safety requirements, the proposed RAID control within the MRNC framework of the SSWMR constrains AI-generated tracking performance within predefined overshoot and steady-state error limits, while ensuring robustness and system stability by compensating for modeling errors, unknown RBF weights, and external forces. Experimental results verify the proposed MRNC framework performance for a 4,836 kg SSWMR operating on soft terrain.
Paper Structure (11 sections, 41 equations, 10 figures, 2 tables)

This paper contains 11 sections, 41 equations, 10 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Lidar-Inertial Slam-Based Navigation
  3. Lidar Odometry
  4. Loop closure: radius search
  5. Design of Control System
  6. Path-Following Control
  7. Inverse Kinematics of SSWMRs
  8. Safety-Oriented RAID Control for In-wheel Actuators
  9. Stability Analysis
  10. Experimental Result Analysis
  11. Conclusion

Figures (10)

  • Figure 1: The whole proposed MRNC framework for SSWMRs.
  • Figure 2: The system structure of LIO-SAM shan2020lio.
  • Figure 3: Impacts of lookahead distance krecht2020possible.
  • Figure 4: The operation under low-light conditions and on soft terrain using the proposed MRNC framework (see Fig. \ref{['schematicwhole']}). The relevant video is available at: https://youtube.com/shorts/0bQO6-SyoO8?feature=share.
  • Figure 5: Mechanical structure of each side of studied MPD.
  • ...and 5 more figures