Jul 2026· 2026 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM)· pp. 1-6· 0 citations· 19 references
Abstract
This work addresses the problem of increasing the temporal resolution of robot positioning in outdoor inspection tasks by leveraging high-frequency inertial measurements. A learning-based approach is proposed to estimate incremental displacement from IMU data combined with GNSS/RTK positioning, using data collected along a predefined trajectory with a mobile robotic platform. Two neural architectures, LSTM and Transformer, are evaluated under different data preparation strategies. Offline validation shows that variations in hyperparameters have limited impact on performance, while the adopted data representation plays a more significant role, with high-resolution IMU–GNSS alignment outperforming feature-based approaches. The selected models were deployed on the robotic platform and tested in the same environment used for data collection, demonstrating real-time operation at the IMU sampling rate and achieving mean errors of approximately 0.140 m and 0.113 m for LSTM and Transformer, respectively. These results indicate that the proposed approach can enhance positioning update rates and support real-time motion estimation in robotic inspection tasks.
It was seen that the global pose of the robot accumulates error and suffers from drift over time but can be improved with an optimization implementation comparing position points to a generated submap which is planned as a future research direction.
M. Peiris, H. Lang, M. El-Gindy et al.· Journal of Physics, Conferen...· 0 citations
This work forms a prior method based on Deep Deterministic Policy Gradient controller and a adaptive unscented Kalman filter which continuously providing constantly estimating robot states and improving the motion primitives in hazardous operating conditions to validate that end-to-end AI navigation architectures deliver the robust performance needed for next-gen automated industrial monitoring and non-destructive evaluation at scale.
Andrey Ershov, Alexey Lyapunov· International Journal of Int...· 0 citations
The problems of partial observability and sensor shortage pose a significant challenge for autonomous Unmanned Aerial Vehicles (UAVs) as they prove to be challenging for conventional Deep Reinforcement Learning (DRL) methods to undertake well under such conditions. In this paper, a memory-augmented Proximal Policy Optimization (PPO) model extended using a Long Short-Term Memory (LSTM) network is proposed as a solution to such challenges. The observation space is constructed from 2D LiDAR and Inertial Measurement Unit (IMU) data to sense simultaneously external observation and internal state of motion, whereas the action space consists of continuous velocity commands. A shaped reward function is optimized for encouraging safe target approaching, obstacle avoidance, and convergence speed. Experimental outcomes show that the PPO-LSTM described herein achieves smoother paths, more robust reward convergence, and a much lower rate of collision than regular PPO. It also generalizes to new environments with movable obstacles. Qualitatively, the success rate increased from 64.5% to 83.9%, collision frequency reduced by over 70%, and path efficiency increased from 0.60 to 0.85, without suffering from unstable training behavior
M. Haddad, Dhayaa Khudher· Kufa journal of Engineering· 0 citations
Reliable localization is required for autonomous mobile robots when individual sensing streams become noisy, intermittent, or unavailable. This study evaluates a multi-sensor fusion framework that combines LiDAR, monocular vision, GPS, UWB, and IMU data using three strategies: (i) a baseline Extended Kalman Filter (EKF); (ii) a dual-stage sequential EKF that refines LiDAR-Inertial Odometry (LIO) before the final fusion stage; and (iii) a hybrid learning-filtering approach in which modality-specific learned motion and position estimates are incorporated into an EKF. All evaluations were conducted in ROS-Gazebo under nominal operation and controlled sensor-degradation/dropout conditions. Relative to the controller-derived reference trajectory, the standard EKF achieved 0.2235 m RMSE and the dual-stage EKF achieved 0.2029 m RMSE, a descriptive reduction of approximately 9.2% for the reported run. The hybrid learning-EKF achieved 0.212 m RMSE under nominal sensing and 0.384 m RMSE during the tested failure sequence. These results support the evaluated fusion designs under the reported simulation conditions, but they do not establish statistical generalization or universal real-world resilience; independent ground truth, repeated trials, GPS ablation, and physical validation remain necessary.
Muhammad Shahzad Alam Khan, Anas Bin Aqeel, Hassan Elahi et al.· Scientific Reports· 0 citations
Simulation and physical experiments confirmed collision-free navigation and successful quick response (QR)-code-based goods inspection, demonstrating the feasibility of the proposed framework for small, structured indoor environments.
T. Q. Le, T. Luu· IAES International Journal o...· 0 citations
Objective: In the field of mobile robotics, autonomous navigation in dynamic environments is one of the most challenging tasks in these environments: traditional methods based on pre-mapping and geometric planning are not effective in these environments due to uncertainty, and reactive methods are lacking in foresight. The work in this thesis tackles these issues by developing and testing an end-to-end Deep Reinforcement Learning (DRL) framework for maples navigation. Methods: A Proximal Policy Optimization (PPO) agent is trained using observations from LiDAR and goal-relative inputs in a high-fidelity open-source simulator that has been domain randomized in order to improve generalization. The trained policy is then transferred to a physical TurtleBot3 platform, with a safety supervisor controlled fine-tuning process. Comprehensive evaluation is performed on five simulated test scenarios (S=5), with 50 episodes per test scenario (250 episodes in total), with real-world trials performed on 20 trials in two different physical settings: cluttered lab and pedestrian corridor. The proposed approach is compared with a DDPG agent, as well as a standard A*+DWA pipeline using paired t-tests (α=0.05) and two-proportion z-tests, with statistical significance confirmed. Results: In simulation, the PPO agent has a success rate of 94% and a normalized path length of 1.18, both of which are significantly higher than those of A+DWA, which are 76% and 1.32 respectively (p<0.001 for both metrics); the agent takes 28.3 s to complete the task, which is significantly faster than A+DWA's 41.2 s (p<0.001). The success rate in the real-time laboratory test is 90%, and the inference time is 8.5ms/s. When the approach is used in structured corridor settings, the A+DWA baseline outperforms the PPO baseline with a 95% success rate vs. 85% for the PPO baseline, though. Novelty: The results of this study show that a PPO-based policy trained only in simulation and fine-tuned only on a small number of real-world tasks can compete with the classical and alternative DRL baselines in unknown and dynamic environments. The results form a feasible basis for the implementation of a learning-based navigation controller on low-cost mobile platforms and offer a fair comparison between the performance and limitations of learning-based navigation controllers and traditional ones.
Nabeel Muhamed, Khaleel Ali Khudhur· Jurnal Media Elektrik· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
