Intelligent Service Robotics最新文献

With the advancement of robotics, mechatronic systems, and automation systems, bilateral teleoperation systems are utilized for performing tasks in remote environments based on commands provided by the master. In application domains like drilling, space operations, medical surgery, undersea exploration, and several other areas, remote task operations are performed using teleoperation systems. Good transparency based on the force feedback and position tracking is still challenging tasks among conventional teleoperation systems. Hence, in order to overcome the challenges, radial basis function neural network (RBFNN) and sliding mode slave teleoperation controller-based disturbance observer (SMSTC-DOB) are proposed in this research. Here, the role of the RBFNN is to estimate the environment parameter for the desired trajectory planning. Besides, the SMSTC-DOB-based slave design helps to synchronize the performance between the slave and master for obtaining stability and good transparency by considering issues like nonlinearities, uncertainties, passivity, and time delay. The implementation is employed in MATLAB/Simulink, which depicts the better transparency of the model in terms of force feedback and position tracking.

Path planning algorithm has always been the core of intelligent robot research; a good path planning algorithm can significantly enhance the efficiency of robots in executing tasks. As the application scenarios for intelligent robots continue to diversify, their adaptability to the environment has become a key focus in current path planning algorithm research. As one of the classic reinforcement learning algorithms, Q-learning (QL) algorithm has its inherent advantages in adapting to the environment, but it also faces various challenges and shortcomings. These issues are primarily centered around suboptimal path planning, slow convergence speed, weak generalization capability and poor obstacle avoidance performance. In order to solve these issues in the QL algorithm, we have carried out the following work. (1) We redesign the reward mechanism of QL algorithm. The traditional Q-learning algorithm’s reward mechanism is simple to implement but lacks directionality. We propose a combined reward mechanism of "static assignment + dynamic adjustment." This mechanism can address the issue of random path selection and ultimately lead to optimal path planning. (2) We redesign the greedy strategy of QL algorithm. In the traditional Q-learning algorithm, the greedy factor in the strategy is either randomly generated or set manually, which limits its applicability to some extent. It is difficult to effectively applied to different physical environments and scenarios, which is the fundamental reason for the poor generalization capability of the algorithm. We propose a dynamic adjustment of the greedy factor, known as the (varepsilon -acc-increasing) greedy strategy, which significantly improves the efficiency of Q-learning algorithm and enhances its generalization capability so that the algorithm has a wider range of application scenarios. (3) We introduce a concept to enhance the algorithm’s obstacle avoidance performance. We design the expansion distance, which pre-sets a "collision buffer" between the obstacle and agent to enhance the algorithm’s obstacle avoidance performance.

In response to challenges faced by mobile robots in global path planning within high-resolution grid maps—such as excessive waypoints, low efficiency, inability to evade random obstacles, and poor maneuverability in narrow passage environments during local path planning—a robot path planning algorithm is proposed. This algorithm integrates state-based decision-making A* algorithm with inertial dynamic window approach. Firstly, the exploration method of the A* algorithm is enhanced to dynamically adapt to the current state of the mobile robot, reducing the number of exploration nodes to improve exploration efficiency. Redundant turning points are eliminated from the original planned path to optimize the global path. Next, a path deviation evaluation function is incorporated into the speed space evaluation function of the dynamic window approach. This function adds weight to forward movement along the original direction, enhancing the robot’s ability to navigate through narrow environments. Finally, key points of the global path are used as sub-goals for local path planning, achieving a fusion of approaches. This enables the robot to simultaneously determine the optimal global path and perform random obstacle avoidance. Experimental verification demonstrates that deploying this integrated algorithm enhances exploration efficiency, reduces path turning points, achieves random obstacle avoidance, and excels in narrow passage environments for mobile robots.

One of the most important aspects in the design of a social robot is its visual appeal, with the design of its head playing a particularly important role in this regard. The head design for social robots has been developed using a variety of ways; one that has become popular today is the use of an in-head projector to create a 3D face for the robot. In this research, we review the design specifications and development stages of the Taban 1 and Taban 2 social robots, which were developed for communication with children in educational sessions. One notable feature of these robots is the presence of a projector located inside the back of the head, which displays the image of different characters on various 3D masks, enhancing the robot's appeal and preventing children from getting bored with the interaction. Due to the low attractiveness of the Taban 1, the Taban 2 robot was developed to increase its desirability. The study explores the conceptual and detailed design of the robots, including their hardware and software components. As children prefer a more cartoon-like horizontal face, this study also highlights the advantages of a horizontal face design, allowing for more cartoon-like characters. To evaluate the effectiveness of child–robot interaction and to study whether the Taban 2 robot is more attractive to children than the Taban 1 or not, acceptance sessions were conducted. The participants expressed high satisfaction and positive reception towards Taban 2, considering it a likable, intelligent, and safe technological teaching aid.

This paper presents the design and analysis of a biomimetic underwater snake-like robot, addressing the main limitations of current underwater robotic systems in terms of maneuverability and adaptability in complex environments. The innovative design incorporates flexible joint modules that significantly enhance the robot’s ability to navigate through narrow and irregular terrains, which is a notable limitation in traditional rigidly connected underwater robots. These flexible joints provide increased degrees of freedom and enable the robot to absorb and release energy, ensuring stability even under external impacts, thus extending the operational lifespan of the robot. Finite element analysis demonstrates the flexible joints’ superior performance in various underwater conditions, offering a greater range of motion and workspace compared to rigid connections. The results indicate that the robot’s modular design, combined with the flexible joint module, leads to improved agility and maneuverability, allowing for precise and intentional operation. The control module, equipped with advanced sensors and a CPU, manages the complex dynamics introduced by the flexible joints, ensuring effective navigation and operation. The specific advantages of this design include the robot’s enhanced structural integrity, its ability to conform to irregular surfaces, and its adaptability to environmental variations. The paper concludes with a discussion on the implications of these findings for the future design and operation of underwater serpentine robots, emphasizing the need for a balance between the effects of elastic modulus and workspace to maximize the benefits of flexible joints.

The use of robots as personal assistants has gained significant interest in recent years. In this research, our motivation is to employ a robot as a personal assistant to optimize the office ergonomics for students. Our personal assistant system consists of DARWIN-OP2 robot, reinforcement algorithm, ROS, communication with robot (using text to speech and speech to text capabilities), and bad posture detection. We conducted a case study on the personal assistant system. The robot receives feedback from student subjects through verbal chatting. Then, the robot executes some tasks such as performing actions or suggesting verbal advice’s to improve the student’s ergonomics. The study included a user evaluation of the robot’s performance, which involved a group of 31 student participants using the robot for a certain period of time. The results show that the DARWIN-OP2 robot is able to effectively and correctly provide valuable health exercises that relieved users’ pains. Additionally, student subjects reported high levels of satisfaction with the robot’s performance and perceived the robot as a helpful personal assistant as it helped in improving their ergonomics. In particular, evaluations of the system, using the group of 31 students, show the system scores 7.7 (out of 10) in speech recognition; 9.7 in health advice’s pain relief; and 9 in users’ opinion on using DARWIN-OP2 as a personal assistant.

Path planning and collision avoidance are vital aspects of successful development and utilization of robots in complex and multi-agent environments. With the integration of robots into social settings, the significance of this issue becomes more apparent. This paper introduces a decentralized management approach based on deep reinforcement learning, where each agent learns independently based on its local observations. The proposed method employs a feature fusion technique which combines 1D, 2D, and 3D features. In order to streamline computation and optimize the training process, an established separation index method is utilized. This approach strategically selects a subset of the most informative features. The presented approach outperforms classical and learning-based methods in various environments with differing densities. Performance evaluation metrics include the interaction index, which indicates the percentage of collision-free scenarios, the reachability index, measuring the time for the slowest agent to reach its goal, the field of view index, demonstrating reduced computation time by narrowing the field of view without compromising interaction, and the scalability index, quantitatively measuring a system’s capability to efficiently handle increasing amounts of work or its ability to be enlarged to accommodate that growth. The performance of this method, compared to PRIMAL, ORCA, and ODRM* methods, has shown an increase of over 30% in situations where the environment is more complex and the number of agents is higher.

Aiming at the shortcomings of traditional A* algorithm in 3D global path planning such as inefficiency and large computation, an A* optimization algorithm based on adaptive expansion convolution is proposed to realize UAV path planning. First, based on the idea of expansion convolution, the traditional A* algorithm is optimized to improve the search efficiency by improving the search step length and reducing the number of nodes needed to select the extended nodes in path planning; adding a weight factor to the cost function to select the appropriate weight of the cost function by keeping the principle of optimal path length while accelerating the planning speed to improve the planning speed of the algorithm; finally, using path pruning to further optimize the paths and reduce the problems of path redundancy. The simulation analysis results show that compared with the traditional A* algorithm, the improved algorithm in this paper reduces the number of extended nodes and shortens the planning time.

This research aims to design controllers based on the hedge-algebras (HA) theory to control differential robots that track reference trajectories. First, the HA-based controller (denoted as HA controller) is synthesized by selecting a suitable qualitative control rule base for the investigated model as a rule-based optimization problem. Then, the optimal HA-based controller (denoted as oHA controller) is established based on the problem of simultaneously optimizing the rule base, the reference interval of variables, and the fuzzy parameters of the variables. Optimization problems aim to minimize the distance between the robot and the reference trajectory. The optimization problems in this study use the Balancing composite motion optimization (BCMO) algorithm. A controller based on fuzzy set theory (denoted as FC controller) with the same parameters as the HA controller is also included for comparison. The simulation results show that the HA and oHA controllers demonstrate many advantages over the FC controller regarding reference trajectory tracking ability, calculation time, and control robustness. The main contribution of this work consists of (i) The development of a novel HA, oHA approaches to control a mobile robot to follow reference trajectories accurately; (ii) Providing optimal global-based BCMO in terms of minimal tracking error with computational efficiency; (iii) The investigation of one control rule base for HA and oHA controllers, which is effective for many different reference orbits; (iv) The development of a robust controller that adapts to the robot’s geometric parameters changes; (v) The proposed controllers have superior performance results compared to controllers based on fuzzy set theory in terms of position error between the robot and the reference trajectory, control action calculation time, and robust ability to change robot parameters.