Accurate three-dimensional (3D) localization is critical for robust human-robot collaboration (HRC) in dynamic indoor environments. However, realizing high-precision localization in complex scenarios still faces challenges such as multipath effects, field-of-view occlusion, etc. To address these limitations, we propose Geo-LSTM, a geometry-constrained long short-term memory (LSTM) framework that integrates ultra- wideband (UWB) sensors, inertial measurement unit (IMU), and barometric pressure (BMP) sensors. First, a Simplified Geometric Localization (SGL) algorithm is proposed, which uses dual-BMP sensors and IMU sensor to obtain precise height information and utilizes the geometric relationships between the UWB tag and anchors to compute an initial location estimate, serving as a priori input for the Geo-LSTM network. This Geo-LSTM algorithm then incorporates multi-source geometric information to extract time-series features from the UWB ranging data and the tag's a priori location, further enhancing 3D localization accuracy. The experimental results from the cluttered indoor environments, including real-world HRC tasks with occlusions, show that the Geo-LSTM algorithm achieves an average 3D localization root mean square error (RMSE) of 0.103 m, representing improvements of 38.60% and 31.20% over the weighted least squares (WLS) method and the range-based LSTM algorithm, respectively. These results demonstrate Geo-LSTM's potential for reliable multi-sensor 3D localization in HRC applications.

In this article, we introduce an assistive device for the elbow joint that is easily wearable, lightweight, and cable driven using a series elastic actuation principle implemented by an endless-shape elastic bungee element. This type of elastic element is selected due to its intrinsic damping, and compliant features similar to human muscles, which provide mechanical filtering against dynamic uncertainties such as impulsive forces, and oscillated movements because of possible controller issues. Furthermore, a spool system is designed targeting to maximize the transmission of the generated elastic force to the wearer while avoiding multiple coiling for the cable wrap/release operation. The design parameters of the device are selected through design and optimization studies. The manufactured actuator's performance is validated on a rigid link under position and force control modes. To demonstrate the effectiveness of the elastic element, we conducted actuator validation tests with and without bungee cases. The results show that while the bungee-integrated setup can transfer the generated elastic force to the link without oscillation at different frequency movements (0.05–0.16 Hz), the bungee-excluded setup performs unstable movements under the same control gains, producing 50.74% more vibrations detected through fast Fourier transform analysis. In addition, to test the system under aggressive conditions, we applied impacts to it, and the results indicate that the damping ratio index of the bungee incorporated system is 56.14% more than that of without bungee case, demonstrating the intrinsic safe behavior of the mechanism. Finally, the device is assessed on six human subjects (different arm weights and dimensions) in a simulated industrial painting task with 5 min duration, with an average <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$22.3^{\circ }/\text{s}$</tex-math></inline-formula> elbow velocity under force control. The average effort reduction on biceps muscle among the subjects is measured 64.42% with respect to the without assistance test.