Abstract In response to the localization challenges posed by multi-modal motion mode transitions during the locomotion of tracked-wheel robots in sheltered spaces, this paper presents a combination localization method based on Inertial Navigation System/Odometry (INS/OD). Leveraging the unique locomotion structure of tracked-wheel robots, a multi-modal odometry motion model and non-holonomic constrained (NHC) velocity model are established. A closed-loop extended Kalman filter is employed to construct the combination localization model, facilitating real-time computation and correction of the robot’s heading angle using the odometry motion model, effectively mitigating system errors and preventing heading divergence. Practical localization tests were conducted on tracked-wheel robots to assess the performance of the proposed method. The results demonstrate that the combination localization method achieves position errors within 0.3578 meters, representing only 0.31% of the total travel distance, with an average error of 0.1611 meters, thus satisfying the practical localization requirements of tracked-wheel robots.
Introduction: Navigating Sheltered Spaces
Navigating disaster zones, collapsed buildings, or underground tunnels requires robots that can adapt to varying terrain. Multimodal robots, which switch between wheels for efficiency and tracks for obstacle traversal, are well-suited for these "sheltered spaces." However, these environments are typically GPS-denied, forcing reliance on internal sensors.
The primary challenge in these spaces is "heading divergence." When a robot switches modes, standard localization algorithms often fail to account for the different physical characteristics and error profiles of wheels versus tracks. Without external references, small estimation errors quickly compound, causing the robot's internal map to drift significantly from reality.
Methodology: The INS/OD Fusion Framework
The proposed solution is a combined localization system that intelligently fuses data from an Inertial Navigation System (INS) and Odometry (OD) using a 15-dimensional closed-loop Extended Kalman Filter (EKF).
Key technical innovations include:
- Multi-modal Odometry Modeling: Separate mathematical models for wheel and track modes ensure accuracy regardless of the locomotion state.
- Automatic Mode Recognition: The system automatically identifies transitions by comparing INS-calculated displacement against raw wheel encoder pulses.
- Non-Holonomic Constraints (NHC): By assuming zero lateral velocity, the model prevents the filter from estimating unrealistic sideways drift.
- Closed-loop Feedback: Unlike open-loop systems, this architecture feeds error estimates back into the navigation equations in real-time, correcting sensor biases and heading drift before they accumulate.
Experimental Validation
The researchers tested the system using a custom tracked-wheel robot equipped with a DETA100R4G-INS and high-precision RTK-GPS for ground truth benchmarking. The test scenario involved a 117-meter journey over 193 seconds, encompassing flat surfaces, mode transitions, and complex, muddy terrain.
The results were significant:
- High Precision: The maximum position error was restricted to 0.3578 meters.
- Relative Accuracy: This error accounts for only 0.31% of the total distance traveled.
- Drift Suppression: The average positioning error was 98.72% lower than that of traditional EKF methods, demonstrating the effectiveness of the closed-loop feedback and multi-modal modeling.
Conclusion and Future Outlook
This combined localization method provides a robust "dead reckoning" foundation for robots operating in satellite-denied environments. While the current model assumes minimal side-slippage (NHC), it successfully bridges the gaps created by motion mode transitions. Future research will focus on integrating this system with SLAM (Simultaneous Localization and Mapping) to enable fully autonomous navigation in the most challenging disaster and industrial environments.
Summary updated