In Japan, the aging population has led to a concerning decline in motor function, particularly evident in activities like driving, where cognitive abilities and motor skills are crucial. This decline has resulted in an increase in accidents involving elderly drivers, emphasizing the need for rehabilitation. Research indicates that tactile and proprioceptive information plays a vital role in motor learning, but its specific impact was previously unclear. To address this, a study by Takagi et al. introduced a virtual spring connection between participants in a steering task, demonstrating the importance of tactile feedback. This study aims to further explore the impact of tactile-mediated interactions on motor learning, focusing on steering wheel operation as a real-life example. Through a target-tracking game, the study seeks to elucidate how tactile feedback influences motor skill acquisition.

To grasp various objects, pushing an elastic body such as a flexible membrane to the object and hold its shape in a deformed state is effective. And to hold object strongly in this grasping method, the shape adaptability must be improved. However, it is difficult for conventional shape-fitting grippers to adapt to the object shape adequately when push it because the stiffness is the same regardless of the curvature direction. In this paper, we propose a method to improve shape adaptability by providing the elastic body with asymmetric stiffness that is positive or negative for the curvature. In addition, we conducted the simulation of the proposed principle, and confirm whether the positive-negative asymmetric stiffness changes the motion of the gripper.

In this paper, the grasping performance was evaluated for a dual-arm robot arm with a jamming gripper. The dual arm robot equipped with a jamming gripper at the end of the 4 degrees of freedam arm has been developed. The gripper and the dual arm enable the robot to grasp an object independent of its position and shape. From the experimental results, using resin beads (particle diameter 0.3mm) as the internal powder, stable gripping force could be obtained regardless of the angle of the gripper. In addition, the robot’s ability to work could be verified by grabbing objects directly below or in front of it.

In this study, we constructed a control method for a robotic system using a motorless robotic hand, proposed in a previous study. Our research group has developed a robotic hand, driven by the rotation of the wrist of the robotic arm, leading to the robotic hand design without the motor. The developed robotic hand mechanism achieves the grasping and wrist rotation motions. These two motions are switched according to the grasping force applied to the fingers of the robotic hand and the operation of the wrist of the robotic arm. Thus, the control method to operate the position and posture of the robotic arm while maintaining the gripper state, i.e., grasping and opening, is required for the robotic system with the motorless robotic hand. This paper presents the control method and the experimental validation.

In this paper, we present a robot finger module that utilizes a slider-crank mechanism to exert pulling force on an elastic thin plate, actuating the joint. The plate, made of phosphor bronze, connects each link and the slider-crank mechanism. The generated pulling force serves as the torque for each joint, exhibiting significant variation with the crank angle of the slider-crank mechanism. This characteristic translates to greater torque production as the finger joint flexes deeper. We conducted several experiments to evaluate the fundamental characteristics of the module, including its movable range and force output.

We present a pneumatically-driven all-silicone needle gripper capable of grasping noodles and granular foods. This gripper consists of a soft membrane and needle-shaped jaws, and is fabricated by molding silicone material. Experimental verification was conducted using spaghetti and chopped green onions.

This paper proposes a vision-based method for estimating the deformation of a soft robotic hand before grasping an object. Due to the softness of a soft robotic hand, it can adapt and fit to the shape of objects, which enables stable grasping of various objects. However, because of the grasping position and object shape, the hand may not grasp an object in a stable manner. Therefore, it is important to calculate whether the hand can adapt to the object shape prior to grasp, which requires estimating the deformation of the hand. In our method, the deformation model of the hand is introduced. Then, calibration is performed using the image to obtain specific parameter of the hand. Finally, based on the deformation model and predicted contact points, the deformation of the entire hand is estimated. An experiment was conducted to evaluate the accuracy of the proposed method.

In recent years, technology has developed to use teleoperated robots to perform missions in hazardous areas and medical surgeries. In this teleoperation, two robots, a leader and a follower, are used in the leader-follower system. The reader is operated by a human operator, and the follower moves in synchronization with the angle information of the reader. In this method, a manipulator-like robot with two arms can reproduce complex movements of a human being. However, the leader must be operated with one hand in order to have dual arms, which is difficult to achieve with conventional manipulators. Therefore, we developed the shape of the hand to enable manipulator operation with one hand.

The efficiency of mobile manipulation tasks has been expected to improve. One way is to take advantage of mobile grasping as like this study. This study first simplifies mobile grasping as two types of grasp action primitives and a moving action primitive and then develops three fully convolutional neural network (FCN) models that predict a static grasp primitive, dynamic grasp primitive, and residual error of moving velocity on visual observation. The development of multiple task-specific FCN models enables the learning of mobile grasping for various shapes at different mobile manipulator velocities. Our experiments on mobile grasping for variously shaped household objects using a mobile manipulator HSR at different moving velocities demonstrated that the proposed method outperforms comparative methods in terms of grasping performance and pick-and-place efficiency.

In this survey, we show that the posture set of an articulated manipulator of a robot can be represented by a Riemannian manifold and its application to a specific motion. The posture of an articulated manipulator is captured by a set of rotational quantities at each joint. By considering the pairs of rotational quantities as coordinates, the set of manipulator's postures can be expressed in terms of coordinates, and by introducing an appropriate metric, the manipulator can be considered as a Riemannian manifold, and the behavior of an articulated manipulator can be expressed in terms of a Riemannian manifold.

Cell production is a very effective production method given the demand for a wide variety of products, but its automation requires robots that can handle a wide range of tasks. In this paper, we focused on the use of the versatile high-speed multi-fingered robot hand for vibration inspection, and considered the influence of the grasping position on the result of inspection by comparing the results of experiments and simulations based on the assumption that the experiments were conducted. From the results, it was confirmed that the grasping position has a significant influence on the inspection results and that there are points that should be considered in vibration inspection using the robot hand.

Elderly and disabled persons often have limited mobility and are unable to receive packages. For this reason, assistive robots have been researched and developed. In recent years, the tasks of assistive robots have been controlled using a digital twin. However, the robot’s automated movements are limited to simple actions such as looking around and gesturing. In order for the elderly to perform complex picking tasks, they must be able to perform picking with simple operations. In this paper, a Head Mount Display (HMD) is used to run to the package and pick it at the touch of a button. This robot is equipped with an RGB-D camera at its hand and can see the object in three dimensions. The grasping points of the box are calculated by matching the point cloud with a box-shaped picking model. Finally, pick-and-place experiments are conducted to demonstrate the effectiveness of the proposed method.

For large object handling, a robot uses both arms and hands. In this paper, we apply a socalled view-based teach and playback approach to a dual-arm robot. In this approach, the robot is required to measure the position of target object. However, it is difficult for the robot to measure the accurate position by using an RGB-D camera for instance. In this paper, therefore, we allow the robot to estimate the grasping positions of both hands without measuring the target position. This grasping position estimator is based on deep neural networks. The robot then modifies the instructed motions by comparing the estimated grasping positions with those of instructed. In the experiments, we show that the robot with the approach through the proposed grasping position estimator is able to reach the hands toward the target object placed at various locations on a workbench.

Performance degradation caused by torsion of the rotation axis is often a problem for articulated robots. A method has been proposed in which the disturbance torque around the motor rotation axis is compensated by the disturbance observer to improve the target value tracking characteristic and the disturbance suppression characteristic. However, if the stiffness of links is low, the control system tends to become unstable. Therefore, in this paper, a frequency shaped axial torsional torque is fed back to a controller to suppress vibration of the end effector. The shaft torsional torque is estimated using the encoder values of the reducer input/output couplings. Validity of the proposed controller is verified by experiments.

The purpose of this project is to design a load torque controller driven by a servo motor with reduction gear. A low-rigidity coupling is inserted between the motor output shaft and the shaft on the load side, and the motor side angle and the load side angle can be measured with encoders. The comparative method using a linear spring constant model for the shaft torsion phenomena caused load torque estimation errors partially. Therefore, I modeled the nonlinearity of the spring constant of the coupling. Simulation confirmed the nonlinear spring constant model identified in the experiment suppressed the tracking error to less than 0.1N.

This paper proposes two trajectory tracking methods for “Discretely Actuated Robotic Manipulators (D-ARMs).” A D-ARM is any member of a class of robotic manipulators powered by actuators that have only discrete positional stable states such as solenoids. One of the most significant kinematic phenomena of D-ARMs is the discreteness of both input range and end-effector frames. The proposing control methods are for two-dimensional (2D) D-ARM with desired end-effector trajectory denoted with the B´ezier curves. A k-means based clustering technique is applied to narrow the range of solution search. A series of simulation results demonstrates the advantages of each control method.

Multi-fingered hands can achieve dexterous manipulation by utilizing tactile sensing feedback. Abundant tactile information is useful specifically for recognizing contact states and plan motions with a synchronous motion of fingers. However, switching dexterous motions with high density tactile sensors is difficult and thus many studies have achieved one-single motion. In this paper, a neural network-based motion switching controller to achieve blind switching manipulation based on tactile feedback. An AE-LSTM model that combined restrictions on loss functions to improve the ability to switch movements and an attention mechanism that modulated attention to modalities required for movement was proposed. The success rate of manipulation for each model in unlearned initial positions and with unlearned objects was investigated. We confirmed that the proposed method exhibited the highest success rate of in-hand manipulation.

Animals exhibit a variety of gait patterns depending on their locomotor speed and energy efficiency. Although various gait patterns have been studied, the mechanism of gait generation has remained unclear. In this study, a one-input, two-output legged robot with differential gears was developed and ground reaction force equalization was performed to realize a stable gait. The developed robot can tolerate angular velocity differences and equalize the torque transmitted to the legs. To study the feasibility of generating a gait based on ground reaction force equalization, we conducted hardware experiments. In addition, energy efficiency was measured and its relationship to autonomous gait generation was investigated. The results showed that the robot can spontaneously generate a gait pattern by equalizing the ground reaction force, and that the energy efficiency is maximized when the robot does so.

Mobile robots are greatly expected to play an active role in rescue work and recovery at disaster sites. In particular, mobile robots had better possesses advanced mobility capabilities in a disaster situation. However, disaster-stricken areas are often uneven terrain with many steps and obstacles, so that robots are essentially able to operate in such environmental conditions. Many existing mobile robots present prescript line symmetric or plane symmetric structure, and are aimed to avoid falling down. In this study, we develop the multi-legged type mobile robot that can operate even if it falls down by making its structure point symmetric. As a point symmetric robot, we apply a regular octahedron shape to the robot configurations, as well as verified symmetrical functions realize locomotion, object grasping, and those control, regardless of its posture.

Various quadruped robots have been developed in recent years. However, in order for quadruped robots to play an active role in our living space, it is necessary to improve their energy efficiency. Therefore, we consider a quadruped robot with a self-weight support mechanism to improve it. This mechanism constrains only the vertical displacement of the toe. In this paper, a simplified constraint mechanism is proposed for an actual quadruped robot and kinematics simulations are performed on the quadruped model.

Biped robots are expected to work instead of humans in the future. However, when a biped robot walks on a rough road, the robot's legs may lose balance as the legs slip or are taken off the ground by the effect of the walking surface. At this time, the robot falls down in the worst case. Therefore, adaptation to the walking surface is an important issue for biped robots to continue stable walking. This study proposes a method to continue walking in an indoor environment when the robots are forced to walk on a surface where there is a possibility of falling. The proposed method is composed of three points. The state of the robot is estimated by combining the estimation of the walking surface and determining the continuous of walking motion, and the robot continues walking by switching to countermeasure motion.

In recent years, research on quadruped robots with high capability for traversing rough terrain has been active. In particular, in order to realize not only locomotion but also autonomous navigation, including recognition of the environment, robots are equipped with many sensors and computers, and the size of the robots themselves are becoming larger and more expensive. Large-sized legged robots are not only expensive, but also difficult to operate with a small number of people. There are few small quadruped robots that are inexpensive and usable for research and development. multi-legged robot experiments are large-scale, making it difficult for a small number of people to conduct RD efficiently. Therefore, we will develop a new useful platform with highly available, low-cost, and small-sized hardware and flexible and extensible software.

This paper proposes a deformable foot mechanism for a semi-passive walking robot. So far, many foot were developed to enhance the performance of walking robots. Those foots are roughly divided into three structures, and those have pros and cons. Therefore, it is expected that the transition between structures enables us to improve the walking performance further. In this paper, we develop a foot mechanism to transit between the flat foot and spherical shape based on the elastic deformation. Though experiments, we verified the utility of the foot deformation.

We have been developing a bipedal robot for the field of disaster relief in Japan. The robot has a tail part for balancing control, high mobility and enhanced carrying abilities for uneven trains. Lower limb of the robot was developed in this study and control property of the tail was evaluated.

A mechanically stable walking motion constructed as a limit cycle, such as in passive dynamic walking, is known for its efficient energy use during movement. The effectiveness of this walking pattern, known as limit cycle walking, has sparked interest, and is expected to be applied to everyday walking on level ground. Additionally, it is known that the shape and characteristics of the foot part of legged robots performing limit cycle walking contribute to the stability of walking motion and movement efficiency. This study aims to develop a foot mechanism inspired by human foot characteristics during two-legged limit cycle walking. Furthermore, the effect of foot characteristics of legged robots on limit cycle walking is experimentally verified, and the improvement of locomotion efficiency of limit cycle walking machines is examined.

We have developed a new leg mechanism that uses a classical link mechanism for the legs, separating the vertical load related to weight retention and the horizontal load related to propulsive force against the aircraft. In this study, we developed a subsystem that performs phase control of one leg by applying the control electrical system for inexpensive electric motorcycles, which have become widely available in recent years, to the legs of the "BigGoat," which have the highest degree of mechanical completion, in view of the characteristics of this leg drive link mechanism that can use a power source for a vehicle. We evaluated the possibility of constructing a subsystem that performs phase control of one leg by applying the control electrical system for low-priced electric motorcycles.

The authors have investigated methods for generating a stable wheel gait of a planar X-shaped walker by rotating two leg frames in the same direction. In a previous work, we achieved generation of a steady wheel gait for a 3-DOF X-shaped walker with two control inputs, in which a reaction wheel is sandwiched between two leg frames. Regardless of the linearization of the model, the rotational motion of the legs could be easily generated using continuous-time output deadbeat control, but it was difficult to determine the initial angular velocity of the reaction wheel that behaves as zero dynamics in the nonlinear model. It was also found that two different steady wheel gaits with short and long periods appear. This paper further analyzes the gait efficiency in more detail using evaluation indices such as step period and specific resistance.

In this study, we proposed the origami-type leg structure with leg lifting ability by driving the base of the leg and utilizing the interlocking property of the origami structures. As an origami structure with interlocking property, we focused on the structure called “Oriceps” the tip of the structure moves back and forth by bending and unfolding the central fold line. We proposed dual half-Oriceps structure based on the Oriceps structure for origami-type leg structure with leg lifting ability. The output angle of each half-Oriceps structure was designed to increase the amplitude of the back-and-forth and up-and-down movements of the legs. The leg lifting of 84.1 mm was achieved by changing the angle of the central fold from 11.4° to 90° on a dual half-Oriceps structure, each with a tip length of 35 mm.

In this study, we investigated the walking characteristics of a passive walking robot with two arms. Specifically, we utilized a simple rimless wheel passive walking robot and mimicked the movement of human arms through symmetrical wobbling masses attached to the body. The simulation results demonstrated the generation of a stable period-1 limit cycle gait. Furthermore, while various trends were observed in walking frequency and speed, it was evident that a larger angle of front-to-back swing of the arms corresponded to an increase in speed. Additionally, not locking the arms allowed for the generation of gait even on very gentle slopes. Nevertheless, it was revealed that improper setting of control parameters could lead to a decrease in walking performance. This study provides valuable insights into the design and control of biped walking robots, contributing to the development of future walking robots.

Tensegrity, as a novel means of connection, is increasingly crucial in the field of robotics. The coupling methods of its individual struts vary widely, and different connection approaches result in distinct dynamic characteristics. However, there hasn’t been in-depth research regarding the essence of its connection methods. This study proposes a novel modeling approach for arbitrary coupling in tensegrity structures based on the rimless wheel model. Through a dynamic model, the generality of coupling forces is elucidated. Furthermore, passive gaits are achieved through numerical simulation, and the soft characteristics are investigated to interpret kinetic energy loss during collisions. This study emphasizes the versatility of the proposed model and provides insights into energy utilization within tensegrity systems. It underscores the importance of understanding and optimizing energy management in tensegrity structures, offering a generalized approach to dynamic modeling for robots.

A lot of research have been done on spherical roller drive according to the type and number of rollers. Typical examples are a sphere-riding robot with three omni-rollers and an omnidirectional wheelchair robot with two normal rollers. On the other hand, in a competition called RoboCup, in which robots play soccer, a sphere is held by driving rollers with constraints. However, there has been no consideration of slippage in the motion of the sphere driven by rollers so far, and there was no mathematical basis for the ball holding mechanism. Therefore, a model with ordinary rollers considering slippage was proposed, and the kinematics for arbitrary placement was established. This paper describes a method for verifying kinematics based on image sensors.

In recent years, double omni-wheels have been widely used in applications such as wheelchairs and automatic guided vehicle in factories. By using free rollers with low rigidity elastic materials, there are benefits such as improved friction and vibration absorption. However, due to the principle of omni-wheels, the contact area between the free rollers and the ground may change during rotation of omni-wheel, causing to vertical vibrations in the vehicle. This paper proposes an optimal design method for omni-wheels to reduce these vibrations. Specifically, it derives the contact area under limited conditions and minimizes changes in the contact area during rotation of omni-wheels. This method was applied to 2x2 double omni-wheels, we achieve the optimal shape.

An articulated mobile robot with an active torsion mechanism can generate the walking gait of a lizard by shifting the phase of its torsion angle and flexion angle. In this study, the effects of the joint positions on walking motion were verified by simulation tests. As a result, it was found that walking performance is better when the torsion joint is at the same height as the wheel axles and the flexion joint is positioned forward of the center between the front and rear axles. In addition, the effectiveness of the walking motion in step-climbing and on a sloping sandy terrain was verified through experiments.

In recent years, 50 years have passed since the period of high economic growth, and due to the collapse of social infrastructure, familiar considerations and differences have arisen. Additionally, due to the increase in natural disasters, it is often difficult to move equipment and operate vehicles on uneven terrain at disaster sites. In this research, we will develop a variable diameter wheel that applies an iris mechanism. This variable-diameter wheel has a simple mechanism that allows the wheel diameter to be increased, decreased, and rotated using two actuators. We will develop and control vehicles with highly traversable mechanisms that can overcome high steps and turn, which are unique to rough terrain environments.

Asymmetric elastic lugged wheels surpass the performance of traditional fixed-lug or lug-less designs, delivering improved traction on sandy landscapes, superior navigation over irregular terrain, and minimized vibration on flat surfaces. This study explores how the inclination angles of these lugs influences overall mobility. Our findings reveal that while changes in lug angle have a minimal impact on traction during forward rotation in sandy conditions, they boost performance during reverse rotation as the angle increases. Additionally, increasing the lug angle enhances the wheel’s ability to overcome obstacles, highlighting the importance of precise lug positioning in maximizing mobility efficiency.

This paper describes the development of a crawler robot capable of traveling in muddy ground. In the pneumatic caisson method, a box-like structure called a caisson is submerged underwater. The working chamber under the caisson is pressurized to prevent the ingress of groundwater, and the ground is excavated. Because this method has significant adverse effects on the human body This method is being promoted to be unmanned by robots. However, robots currently in operation have the problem that the ground at the work site is uneven. To solve this problem, a crawler robot that can travel over muddy ground without getting stuck is needed.

In recent years, there has been much anticipation for the practical application of robots that provide services in various fields. The Robotics Laboratory of Kobe City College of Technology has been developing "Kobeport(oo)n," a two-wheel drive autonomous mobile robot, with the goal of performing park manual tasks in place of humans. In demonstration tests, the robot could not climb steps and slopes present in the park. In this study, an experiment was conducted to improve the robot's ability to run over irregular terrain, and we proposed as well an evaluation method to grade how well the robot did on this rough terrain.

Wheels with free rollers, which are often used as omni-directional moving mechanisms, have problems such as vibration during progression to the wheel's rotation axis, and a large difference between the large wheel diameter and the free roller diameter, which results in extremely low tread performance depending on the direction. To solve these problems, this study proposes a mechanism in which the passive wheel is mounted at an angle so that the axis of rotation of the free roller of the mechanum wheel is perpendicular to the main direction of travel.

This mounting can significantly reduce vibration when moving in the main travel direction without losing the advantages of conventional wheels with free rollers, and also enables the wheel to step over bumps not only in the front-back direction but also to some extent in the left-right turning direction. The fact that We have confirmed that.

This paper presents an implementation of an active-caster type omnidirectional robot using planetary gear as steering mechanism, which distributes two motor input via the sun gear and the internal gear into steering as the carrier motion and wheel-driving output as rotation of a planetary gear on the carrier. The idea cited from previous study is shown first with our additional idea to drive the sun and the internal gear, which makes the caster unit compact. A set of control method of caster unit and whole robot light enough to be implemented on a 16bit microcontroller with fixed-point calculation is provided with experimental demonstration in accompanied video.

We have been studying a mechanism to realize stable mobility without burial by efficiently generating buoyancy and forward force in snowy and icy environments and its performance verification in non-natural snowy environments. The mechanism proposed in the previous paper was mainly intended to secure "buoyancy" against snow. However, it caused the collapse of the surrounding snow because it compacted the snow in a small area. Therefore, there was a risk of burial in fresh snow. In this report, we propose a mechanism called a crawler-mounted passive wing wheel. This mechanism can simultaneously provide flotation force and forward force. At the same time, the mechanism transforms so that the rotating legs are placed at a deeper angle in the snow. This action allows the robot to escape from the snow. In the previous paper, we proposed a pseudo-snow made of a mixture of plastic pipes. In this report, we focus on powdered materials as pseudo-snow. By adding water to these materials, the physical characteristics of snow are reproduced.

In this paper, we design a cost for entering a slope in a positive direction and evaluate automated operation of dump trucks with a path planning method using that cost. We analyzed data from human-driven dump trucks. We use the tendencies identified by this analysis to design the cost of slopes. We conducted an automatic operation of a dump truck in a quarry to evaluate the inclination of the roll axis. Experimental results showed that when slope direction was taken into account, the maximum tilt of the roll axis was smaller than when conventional Hybrid A* was applied. As a result, it was confirmed that safe automatic operation is possible even at the quarry site with slopes.

Investigating the deterioration of buildings on Gunkanjima, which is a component of World Heritage, is crucial for preserving the island’s warship-like ridge and safeguarding the Industrial Revolution Heritage of Meiji Japan for future generations. In FY2023, the Committee for the Evaluation of Building Deterioration Status and Structural Strength is conducting a comprehensive survey of all buildings on Gunkanjima. However, concrete structures over 100 years old, such as Building No. 30, have deteriorated to the extent that human interior surveys are no longer feasible. This paper presents the findings from an investigation into the deterioration of Building No. 30, which is inaccessible to humans, conducted using a tracked vehicle equipped with a work arm. This represents the world’s first attempt at employing a tracked vehicle with a work arm for the investigation and inspection of the interior of a building where human access is restricted.

Some organizations have planned to explore the Lunar or planetary surfaces using a small rover. Additionally, exploration using a wheeled swarm rover has been planned. However, the traveling performance of the single rover could be worse due to loose soil on the planetary/Moon surfaces. To solve this problem, our previous study developed collaborative locomotion via an inching locomotion. Although the proposed locomotion suppressed the slip by increasing connection rovers, the traveling performance of the method when the high load is applied has remained unclear. This paper investigates the relationship between the load and the traveling performance when the connection rovers are changed. Experimental results further clarified the suppressing effect by increasing the number of rovers.

Simulators and routing tools are important in the development of autonomous mobile robots to check their operation. Many researchers have developed their own simulators or use ROS, each of which has its own advantages, but there are also problems such as complexity of development and limitations of use. The objective of this research is to develop a simulator for a robot moving in three-dimensional space that operates in real time, without relying on existing robot development middleware. We aimed to develop a system that allows easy path setting on a GUI and a system that can control a mobile robot in a physical simulation to enable more efficient robot development and research.

This paper describes a mobile robot equipped with 4 bar linkage mechanism designed to climb step obstacles. To balance turning ability and step overcoming capability is the difficult problem to design a mobile robot. First, the high maneuverability of the multi-articulated crawler robot is confirmed through mechanical analysis. Next, an idea of applying 4 bar linkage to a passive crawler unit is confirmed. Finally, the effectiveness of the proposed mechanism is confirmed through an experiment with a prototype robot.

With the recent expansion of the e-commerce market and the decrease in the labor force, the demand for mobile robots has been increasing and various studies have been actively conducted. We have combined a 3D LiDAR-based scan matching technology with an active-caster type holonomic omni-directional mobile robot that can generate arbitrary speed and angular velocity at any time and in any direction with high precision, aiming to realize advanced and autonomous omni-directional mobility indoors with high accuracy. As a result, we succeeded in (1) efficiently obtaining high-quality maps even in a narrow driving range, (2) controlling the robot's movement using the results of 3D LiDAR-based localization, and (3) performing autonomous avoidance movement against known obstacles.

In recent years, automation is being promoted in the medical field. However, conventional research has problems such as being costly and Using robots have difficulty moving left and right and changing direction. To solve these problems, we propose Omni-directional coordinated conveyance system using mobile robots. This system can be freely connected and disconnected from the transfer object. In this study, corner point detection is performed using information on the transfer object acquired by 2D LiDAR and distance differentiation data between the transfer robot and the transfer object. Then, by performing corner point matching with Multi-step Minimum Error Search, estimates of the robot’s relative position and direction are output. Experiments confirmed that it is possible to output estimates of the robot's relative position and direction when scanning the cart by performing corner point matching using Multi-step Minimum Error Search. We also confirmed issues with the output results.

Various studies have been conducted on self-localization to enable mobile robots to move freely within facilities. However, adapting to dynamic changes within the facility poses a challenge. In recent years, there has been a growing interest in initiatives like ”robot-friendly” environments, which aim to facilitate robot activities. This study achieves robust self-localization by installing markers in the environment to aid the robot in acquiring its position accurately. Since obtaining accurate self-localization with a single marker is challenging, this study develops an algorithm that utilizes two markers to estimate position accurately. Using this algorithm, it was confirmed that the mobile robot can enter the elevator from the corridor without colliding with the walls.

This paper proposes a path planning method integrating the A* and ORCA algorithms to enhance the obstacle avoidance capabilities of autonomous mobile robots. It implements efficient path planning around static obstacles using the A* algorithm and avoids collisions with moving obstacles through the application of the ORCA algorithm. We have developed a system that performs path planning using these algorithms, utilizing Unity for development. The system employs Laser Range Scanner to detect real-world obstacles, incorporating this data into the simulation for path planning. Our aim is to evaluate the system's effectiveness through both simulations in Unity and real-world robot experiments.

This paper proposes an algorithm to maximize the number of inspections conducted by a robot equipped with a THz sensor in dynamically crowded environments for security screening purposes. This algorithm represents the movement of robots and individuals using time-space networks and solves the shortest path problem using Dijkstra’s algorithm to maximize the number of inspections. The effect of pedestrians’ route changes simulated by the social force model was verified through simulations, revealing a decrease of approximately 6%. Additionally, an algorithm considering pedestrians’ route changes was proposed, and simulation results demonstrated that the proposed algorithm could inspect approximately 6% more pedestrians.