Mobile Robot Localization Based on FEM Stress Analysis Using Pressure Sensors under Floor

Traditional mobile robot localization techniques relying on on-board sensors often face significant limitations, such as visual occlusion, overreliance on visual features, and high computational costs. To address these challenges and monitor the entire environment in real time, this paper proposes an innovative localization framework that uses pressure sensors embedded under the floor to estimate a mobile robot pose based on its weight distribution. In order to reduce computational complexity, a mathematical model based on the Kirchhoff–Love plate theory is established to describe the ground deformation under external loads. This model is then used to simulate and analyze the stress distribution using the finite element method (FEM), forming the basis for the localization. The robot pose is iteratively estimated using a particle filter-based approach, which dynamically adjusts based on observed and predicted pressure distributions to arrive at the optimal result. Using environmental feedback rather than relying on on-board sensors, our approach eliminates the need to equip robots with dedicated localization hardware, reducing cost and system complexity. I. INTRODUTION Accurate localization is one of the most critical compo- nents in mobile robotics, as it allows robots to understand their pose within an environment and navigate effectively [1]–[3]. Over the years, researchers have developed a variety of indoor-based localization methods, including vision-based simultaneous localization and mapping (V-SLAM) [4], [5], and light detection and ranging (LiDAR) technology [6], [7], highlighting their applications, challenges, and opportunities for improvement. In terms of V-SLAM technology, cameras capture detailed visual information, enabling robots to iden- tify and differentiate between features such as walls, doors, and objects. The V-SLAM relies on clear and consistent views to detect and track features in image data. If the field of view of the camera is obstructed or the environment lacks distinct visual features (e.g., plain walls or open spaces), the algorithm may not be able to locate the robot [5], [8]. Similar problems also occur with LiDAR. LiDAR provides precise distance measurements, enabling robots to generate highly detailed maps of their surroundings. However, LiDAR may struggle to detect or accurately represent transparent obstacles, such as glasses or acrylic boards [7]. As an alternative approach to localization, we aim to find a novel localization method that relies on environmental sensors instead of on-board sensors. Pressure sensors, as a commonly used medium for robots to sense the external envi- ronment, are widely used in various fields such as industrial All authors are with Graduate School of Advanced Science and Tech- nology, Japan Advanced Institute of Science and Technology, Japan. {s2520026, ji-y}@jaist.ac.jp Fig. 1. Localization system using pressure sensors under the floor. robotic arms and human-computer interaction [9]–[11]. [9] compared the effect of the pressure sensor measurement on the ground reaction force by investigating the accuracy of the measurement under standard loading conditions. A notable study by [10] demonstrated a method in which sensors installed on the robot’s foot captured terrain characteristics, aiding in surface classification. However, these approaches face significant challenges in localization tasks because of their dependence on pressure distribution in the area near the robot and lack recognition of the global environment. On the other hand, commercially available systems such as pressure sensing mats from Tekscan Inc. have shown great promise in capturing detailed force distribution data [12]. However, their design prioritizes high-precision biomechanical and industrial use cases by using a huge array of sensors, making them less efficient and scalable for robotic localization tasks in dynamic or expansive environments. We hope that by arranging fewer pressure sensors, the sensor system can be applied to robot localization in a wide range of indoor environments. In this paper, from the perspective of pre-establishing an environment for localization rather than for each robot, we suggest a novel concept to estimate a robot pose using pres- sure sensors installed under the floor as shown in Figure 1. We measure the pressure distribution of the robot on the ground by arranging a sensor matrix under the floor. The basic idea is to estimate the location of each load point, considering the bending problem of the thin plate due to the load of each wheel. Inferring the location of the load using the values of the pressure sensor matrix is a nonlinear problem; thus, the particle filter-based approach is applied 2026 IEEE/SICE International Symposium on System Integration (SII) January 11-14, 2026. Cancún, México 978-1-6654-5784-2/26/$31.00 ©2026 IEEE 921