
With their revolutionary applications and potential for innovation across a wide range of industries, artificial intelligence (AI) and robotics are expected to continue influencing society and business. This study aimed to review recent literature published between 2018 and 2026 in the areas of AI and Robotics. The “Preferred Reporting Items for Systematic reviews and Meta-Analyses” (PRISMA) guidelines are used to conduct the research in the area of AI and robotics. The search strategy resulted in the selection of 119 studies based on the exclusion and inclusion criteria. The literature review revealed that robotics and AI are widely used across manufacturing, logistics, agriculture, e-commerce, finance, and entertainment to automate tedious tasks, increase production, and improve efficiency. While autonomous drones are utilized for tasks such as transportation, inspection, and surveillance, collaborative robots operate alongside humans in production contexts. Despite significant improvements, issues with robots and AI include data security and privacy, resilience against hostile attacks, the interpretability of AI models, and public acceptance of automation. Stakeholder participation, teamwork, and interdisciplinary research are necessary to address these issues. Some future research directions in AI and robotics include explanatory AI, ethical AI, human-centered design, multi-agent systems, bioinspired robotics, sustainable and resilient AI, and multi-agent systems. Future advancements in these areas could improve the capabilities, safety, and societal effects of intelligent systems.
The convergence of computer vision and motion planning remains a fundamental challenge in autonomous systems, particularly when precision, adaptability, and efficiency are of great importance. Addressing these limitations, we introduce a novel cross-modal intelligence framework that bridges the perception-action gap through two complementary innovations: HFC-YOLOv11 and DE-MPC. The Hybrid Feature Correlation architecture (HFC-YOLOv11) extends traditional object detection paradigms by orchestrating multi-stream detection pathways with a specialized correlation mechanism, creating robust representations that maintain integrity across diverse environmental contexts. Complementing this perceptual advancement, the Differential Evolution-Model Predictive Control (DE-MPC) framework improves motion planning by synthesizing B-spline trajectory parameterization with metaheuristic optimization, enabling fluid and efficient physical interactions. To demonstrate the framework’s viability, we applied it to the autonomous solving of Rubik’s Cube, a notoriously complex 3D combinatorial puzzle, where the system exhibited strong robustness-achieving 98
Accurate real-time detection of the lower limb gait phase is imperative to the safe and effective control of robotic exoskeletons, especially those used by post-stroke individuals with hemiparetic gait patterns; yet, current data science algorithms that are supervised-based are highly dependent on large amounts of labeled training data acquired from healthy subjects, thereby unable to generalize well to pathologies, while fully supervised deep learning models need more than 90
Precise localisation and mapping are essential for reliable sewer inspection. A robust positioning method is required for the deployment of autonomous robotic systems, the detection, documentation and monitoring of defects and the preparation of maintenance work. Pose estimation based on wheel encoders and inertial measurement units suffers from slip and drift over long distances. Standard lidar- and vision-based simultaneous localisation and mapping algorithms, optimised for indoor or underground environments, struggle in sewers because of sparse geometric features, non-planar walls and repetitive structures. To address these challenges, a lidar-based simultaneous localisation and mapping algorithm designed for sewer systems is proposed. The method incorporates domain-specific constraints: pipe segments are straight elements with known or bounded length, starting and ending at well-defined connections. In addition, an adapted feature detector and loop closure strategy are introduced to deal with smooth surfaces, lack of distinct edges and the presence of repetitive pipe sections. By combining knowledge of the geometric properties of the sewer infrastructure with sensor data, the proposed approach achieves stable localisation in laboratory and operational sewer holdings. In the laboratory setup, the method achieves a mean trajectory error of 13.4 ± 4.1 mm . In the field, endpoint errors below one per cent of the holding length are achieved. This level of accuracy provides a pose reference for associating inspection findings with pipe coordinates and supports autonomous sewer inspection. The algorithm closes a technological gap and contributes towards practically deployable robotic tools for large-scale condition assessment, targeted maintenance and long-term monitoring of critical underground infrastructure.
In time-critical missions such as cooperative strikes and synchronized operations, multiple UAV systems demand exceptional path-tracking accuracy and temporal coordination. To this end, under the assumption of fixed-altitude flight, this paper investigates the time-coordinated tracking control problem for multiple UAVs in the lateral-planar space, and proposes a consensus-based distributed coordination control method. Under the virtual target guidance framework, a kinematic model of path-tracking errors is established, and a sliding-mode path-tracking strategy incorporating a nonlinear approach angle function is developed to ensure stable tracking of the reference path for each UAV. The asymptotic convergence of the tracking errors is rigorously established within the Lyapunov stability framework. Building on this foundation, a coordination variable that characterizes task progress, together with a dynamic adjustment mechanism, is introduced to achieve consistency in mission progress among multiple UAVs under time-varying communication topologies. The exponential convergence of the coordination errors is further proven. Simulation results demonstrate that the proposed method achieves high-precision path tracking and temporal synchronization under the considered simulation scenarios. Compared with existing methods, it provides faster coordination error convergence and improved steady-state coordination accuracy, thereby demonstrating its effectiveness for distributed time-coordinated tracking control.