In Japan, the aging population is progressing rapidly. Along with that, lack of care facilities and carers is one of unignorable problems. Despite the fact that the production age group is decreasing, in order to solve this problem, it is not possible to assign many production age groups to nursing care sites. It is necessary to reduce the necessary manpower by technology intervention. Therefore, in this research, we will develop a device for excretion detection using RFID tags targeting care recipients. The goal is to detect excretion by attaching an RFID tag to a diaper and causing it to fail due to urine and feces moisture.

Currently, many pet robots are being researched and developed. In this study, we will study pet robot used to improve dementia. As a result of previous research, this pet robot may be applicable to reminiscence therapy. Pet robots have disadvantages that tend to get bored due to monotonic operation. In the previous research, there were many opinions that the variation of the movement of the robot was small. Therefore, it is necessary to increase the variation of the movement of the robot. We will improve the mechanism of the pet robot to increase the operation. We conduct questionnaire survey and compare with previous research.

In this study, we were aiming to develop a totally new type of power prosthetic hand, which can be equipped with both light and convenient characteristics at the same time. In terms of the actuators design, we all finally chose to adopt the SMA in the study, which has a higher output weight ratio than other actuators, making efforts to try to realize the goal of creating a power prosthetic hand with the light and small convenience that can be hidden in the palm of the hand through the round-trip mechanical movement of the SMA.

In previous study, I developed a forceps force limiter which uses leaf spring buckling. However, the maximum stress of the buckled leaf spring might exceed the allowable stress because this mechanism does not limit the displacement of the leaf spring. This paper proposes a lock mechanism of the force limiter which applies the friction mechanisms of ”Jizaikagi”. The shape of ”Jizaikagi” is slender, which enables to install it in forceps. If the displacement of the buckled leaf spring exceeds the threshold value, then this mechanism locks the handles of the proximal side. The developed lock mechanism has 12 mm in thickness, and I confirmed that the proposed mechanism locked the handles when grasping force was 13 N.

One of the main advantages of surgical endoscopy is that it is minimally invasive. However, since conventional surgical endoscope are rigid, it may damage organs, and it is challenging to reach a target located at the back of an organ. In this research, we have developed a steering mechanism that is flexible and bendable for a surgical endoscope. By applying air pressure, the mechanism can bend in any direction. In this report, to enhance the operability, two functions have been realized: radius of curvature of the mechanism can be varied widely by inserting a thin rod, and the mechanism rotates around its center axis with keeping the tip position in order to rotate the camera image. By the fundamental experiments, we confirmed that the radius of curvature can be changed by the thin rod and camera image can be adjustable.

By integrating locally operated small surgical robots in a sterilized area, a surgeon can perform safe and accurate robotically assisted laparoscopic surgery. At present, there is no locally operated forceps manipulator that can operate within a small space while providing a wide working area on the abdominal wall. In the present study, a new wire-tube driven 3-DOFs forceps manipulator based on the polar coordinate system with two circular ring guided rails for the yaw and pitch axes, and a linear guide roller for the insertion axis though into the forceps that can act as a third arm for the surgeon was designed. The dimensions of the manipulator are 160 mm in outer diameter * 180 mm in height. The operating range is ± 180° for the yaw axis, ± 45° for the pitch axis and 150 mm for the insertion axis.

A new locally operated unilateral master-slave manipulator for endoscopic surgery performed by a surgeon in a sterilized area was developed. The manipulator system consists of a portable master device and a forceps slave manipulator LODEM as the third arm. We propose the 3DOFs portable operating device with gimbals for the pitch and yaw axes, and a linear guide roller for the insertion axis though into the commercial forceps. The dimensions of the mechanical component are 90 mm × 180 mm × 120 mm, and its mass is 470 g. The operating ranges for these axes are ±60 °, ±60 °, and 200 mm, respectively. The mechanical backlash for these axes are 1.7 °, 1.4 °, and 5.2 mm, respectively. The mobile LODEM is unilateral master-slave point to point controlled by the master device. The master-slave manipulator was confirmed for these axes.

By integrating locally operated small surgical robots in a sterilized area, a surgeon can perform safe and accurate robotically assisted laparoscopic colorectal surgery. At present, there is no locally operated forceps manipulator that can operate within a small space while providing a wide working area on the abdominal wall. In the present study, a new 5-DOFs forceps manipulators which is selectable controlled by a handy remote controller that can act as a third arm for the surgeon was designed. The dimensions of the manipulator are 345 mm * 380 mm * 150 mm. The operating range is 0 ° to 90 ° for the pitch axis, ± 45 ° for the yaw axis, 200 mm for the insertion axis, 25 ° for the grasp axis and free rotation for the roll axis.

A new locally operated master-slave multiple connection system for endoscopic surgery performed by a surgeon in a sterilized area was developed. The system can produce selectable connection for portable master devices and assistant slave robots as the third arm. We propose master-slave system between a forceps handle shaped operating device and a forceps robot LODEM which is selectable for Mobile LODEM or Guiding LODEM using a motion control board installed with middleware ORiN-based medical robot architecture MRLink. Furthermore, we propose on/off system between the operating device and an endoscope robot AESOP. Providers for these devices were constructed, and the master-slave system using robot integrator application connected to these providers was confirmed.

A new image processing method using luminance and hue of L*a*b* color space for real time processing and detecting achromatic surgical instrument in endoscopic view has been developed. The proposed method could be used for semi-automatic control of a forceps robot. It makes use of a luminance gamma correction, a masking process based on combination of removal of the halation and the shade using L* axis, the organ’s red color and the fat’s yellow color using a* and b* axes, and the boundary and small area. The area of the surgical instruments recognized in three laparoscopic surgical videos using this method were performed, and the recognition rate was 71 % for the forceps. The average processing time was 14 ms.

A remote-controlled robot for CT-guided needle insertion has been developed to prevent a doctor from radiation exposure. However, it is difficult for a doctor to recognize abnormal states of insertion when the doctor uses the robot because the doctor cannot obtain force sense information from the needle directly. Such an abnormal state occurs when a needle contacts with a bone. Therefore, we propose detecting abnormal states of insertion if force sensor value exceed a threshold and stopping the robot automatically. First, we conducted experiments to determine thresholds. We measured force values when a doctor feels the sign of plastic deformation in needles. In consideration of the stop time delay, thresholds were set by subtracting the rise values of force from the measured force values. Then, we verified that the robot can stop appropriately by the thresholds without occurring plastic deformation.

Deep vein thrombosis (DVT) occurs when a blood clot forms in a lower limb due to a prolonged period of inactivity. By improving the deep vein circulation through ankle exercises, it is possible to prevent DVT. However, current rehabilitation measures, such as manual physical therapy, are labor intensive and time consuming. The existing automatic exercise assisting devices are expensive and bulky. Hence, in this study, we designed a novel wearable DVT prevention device using thin McKibben muscles. Without a rigid structure, the device was composed of only soft materials, was extremely lightweight, and occupied little space around the leg. Ankle exercises in all directions, including dorsiflexion-plantarflexion, inversion-eversion, and abduction-adduction, could be achieved using the proposed DVT prevention device. The range of motion by this device was evaluated. The device achieved 60% of the range of motion for dorsi-plantarflexion, 95% for inversion-eversion and 80% for abduction-adduction.

Endoscopic sinus surgery (ESS) is the mainstream of modern surgical therapy of sinusitis. ESS requires frequent training doe to the complexity of the operation. As for the method of ESS training, inexpensive and simple one is desired because ESS is a solo-surgery, important organs are close, and so on. Terms of this surgical method are simple implementation and quantitative evaluation without using large external apparatus. In this paper, we acquire information by self-position estimation from endoscopic image using ORB-SLAM instead of the 3D-position measuring device. As a result, it’s suggested that the evaluation using the coordinates acquired from the endoscopic image provides the same analytical capability as in the case of using the conventional large-scale apparatus.

This paper presents a surgical robot mechanism optimization method considering of the working error and operator’s muscle burden. The virtual surgical system was developed to simulate the visual and haptic feedback. The participants operated the VR surgical simulation system while the authors measured the working error and the participant’s joint motion. The histogram was made of the data to show the distribution. The authors estimated the appropriate probabilistic distribution model using Akaike's Information Criterion (AIC) method. As a result, there were many cases most applicable to the Weibull model.

Water-Filled Laparo-Endoscopic Surgery (WaFLES) is a surgical technique which is performed by irrigating the abdominal cavity with isotonic liquid. Since saline is used as perfusion fluid in WaFLES, the use of monopolar electrodes is limited. So design of devices for hemostasis in WaFLES environment is warranted. Therefore, we introduced a laser surgical robot system that enables a high precision laser irradiation for WaFLES. In this study, we optimized the laser power, scanning speed and scanning path of the laser surgical robot for automatic hemostasis and evaluated thermal damage and treatment time. As the result, we demonstrated the feasibility of a coagulation method that balances thermal damage and treatment time.

This paper describes evaluation results of a magnetically driven cytology brush applicable to a capsule-type medical device. The cytology mechanism is composed of a nut with a brush and bolt with a permanent magnet. When rotating magnetic fields are applied, the rotation of bolt due to magnetic torque is converted into the linear movement of nut. The periodic switching of the rotational direction of magnetic fields enables a reciprocating motion of the cytology brush. In order to improve cell collection by brushing, a small magnet was attached to the brush tip. The brush applied with a rotational vibration could enhance the performance of cytology.

This paper proposes a novel capsule-type medical device for stomach inspection. The capsule has a swimming mechanism composed of combined permanent magnets and an elastic film. External magnetic field can drive the swimming mechanism and navigate the capsule to the target position of the stomach. In this study, three kinds of operation modes have been evaluated. The first is a straight movement mode in which the capsule moves in the horizontal direction by applying AC magnetic field. The second is a tilting mode in which the capsule tilts in any direction by applying DC magnetic field. The third is a sinking mode in which the capsule can move in the vertical direction by applying AC magnetic field biased with DC magnetic field. The evaluation results suggest that proposed method can be applicable to the stomach inspection.

ESS (Endoscopic Sinus Surgery) requires advanced skills. Therefore, supporting surgeons in improving surgical skills is important. Hence, process analysis in production engineering need to be applied to complicated surgery such as ESS. In the previous research, ESS skill evaluation system using process analysis method was developed which enables detailed evaluation of surgeon skills per surgical process. However, the system requires declaration of surgeons and manual record of surgical process, so they have problem of interruption of surgery and manpower. Therefore, estimating the ESS process automatically is necessary for the ESS skill evaluation system. In this study, we propose the method of ESS process estimation by SVM (Support Vector Machine) based on CT data of patient and surgical tool trajectory data. This methods are evaluated in accuracy of process estimation by clinical data.

In this study, we develop a pneumatic surgical robotic arm not requiring positioning a RCM(Remote center of motion) of the robotic arm on an insertion port of a patient. We use a joint that do not have stiffness but have ability to output torque. We call this joint “semi-active joint”, and realize them with a pneumatic actuator with high backdrivability. Using the semi-active joint, the posture of the robot is formed passively by a constraint of the insertion port. Plus, the robotic arm can be expected to compensate external forces in abdomen. We controlled its end point in a constraint of the insertion port, and evaluated about accuracy of its movement and forces exerted to the insertion port.

Endoscopic sinus surgery （ESS） is now widely accepted in the field of otolaryngology. However, for doctors, routine and repetitive training is indispensable for safe surgery. Therefore, it is necessary to develop an ESS training system having quantitative and objective evaluation that is easy to use every day. In order to evaluate ESS training, we developed a sensorless forceps motion analysis method. In this study, we adopted the deep learning method for extracting the forceps region from endoscopic videos. As a result of analyzing the behavioral features, significant differences were observed between experts and novices in the several proposed indexes, and the effectiveness of the proposed method was shown.

This paper describes a new stent for cerebral catheter operation using electrospinning. In the previous study, we proposed a fractal stent with valve mechanism. In this research, we use electrospinning for fabrication of the fractal stent. We propose a patterning of electrospinning using fabrication of MEMS process. We measure the width of the patterning by electrospinning. In this paper, we show the evaluation of the electrospinning patterning, simulation of thee flow and the fabricated fractal pattern.

This paper examines the feasibility of automated alignment in prostate targeted biopsy by comparing the prostate contour between different modalities. The prostate targeted biopsy that is attracting attention in the treatment of prostate cancer largely depends on the doctor who operates surgery, so it can be expected to reduce the variation in the diagnostic performance by automation. In the proposed method, segmentation is performed using deep learning, and the same prostate cross section between different modalities is estimated from the similarity obtained by comparing prostate contours of different modalities obtained by segmentation. In this method it was possible to estimate close to expert judgment with accuracy of 69.4%. Furthermore, by considering the deformity of the prostate gland and calculating the similarity for each angle, we achieved an estimate close to the judgment of experts with higher accuracy of 83.3%.

Hepatic hemangioma is lesion frequently found in ultrasonography. Although hepatic hemangioma is benign tumor, other tumors are subject to detailed examination by judging hepatic hemangioma. Therefore, in this study, we proposed a new discrimination method aimed at improving accuracy by using deep learning to classify hepatic hemangioma and blood vessels appearing on ultrasound images. This is prevents feature loss system using learning model that removes diaphragm and kidney, sorts out image features to increase the proportion of classification targets. In the experiment of this study, 132 ultrasound images were divided into 120 Training images and 12 Test images. As a result, it was possible to estimate with accuracy of 89.4% in classification of hepatic hemangioma and 79.7% in blood vessel classification by the proposed method.

Bubble removal is indispensable in the priming of the artificial heart-lung circuit. This bubble removing operation is important to remove bubbles quickly and accurately, and in general it is a heavy burden on the person in charge to manually perform. Therefore, in order to reduce this burden, we have been studying bubble removal devices. In this study, the amplitude of the mechanical vibration of the arterial filter was changed, and the effect at each amplitude was investigate：ed as efficient bubble elimination was obtained by applying constant vibration to the arterial filter using this device. We conducted experiments to change the amplitude of the arterial filter in three steps. Three kinds of amplitude of 10 °,20 °,30 ° were given, and a vibration with a constant frequency of 10 Hz was given for 10 seconds. As a result of the experiment, bubbles could be eliminated satisfactorily at an amplitude of 30 °. However, at an amplitude of 20 °,10 °, minute bubbles remained. In the future, we will investigate the influence of the vibration cycle on bubble removal in the bubble removing device.

In respiratory diseases typified by lung cancer, biopsy in the lung using an endoscope is considered as effective for diagnosis. However, even skilled physicians are difficult to insert it. In order to simplify these biopsies, this paper presents a novel operation of a self-propelled catheter and its basic configuration for automatically reaching the target lung periphery. The proposed device consists of a flexible tube with an outer diameter of 2.6 mm, and both of the propulsion motion and steering motion are generated by the pneumatics. We verify the validity of the proposed methods through simple experiments by the developed prototype.

We have been discussing a disaster information gathering system using mobile robot. There is a problem that an accident occurs on the robot in case of disconnecting the wireless LAN between the operator and the robot. This paper describes a method to solve this problem by using LoRaWAN. In a wireless LAN tele-operation, the operator operates the mobile robot by using picture or movie information from the robot. However, in a tele-operation using LoRaWAN, operator can’t get that information by the low transmission rate. Then, we proposed a single-packet tele-operation method different from the conventional tele-operation method. In experimental results, we confirmed that the proposed method is effective for the tele-operation of mobile robot.

In order for humans and robots to comfortably share everyday living space, it is important that we can understand the behavior of the robots. In this paper, we propose a robot that interacts with an opposing pedestrian by moving the face direction. The robot points the face to the pedestrian to show that he has noticed the pedestrian. Then the robot turns the face in the avoidance direction to show the side that he avoids. Subjective evaluation using a questionnaire showed that the pedestrians felt the proposed behavior of the robot significantly more comfortable to pass by and more secure.

The authors have been developing an information-gathering system, Robot Wireless Sensor Networks (RWSN), using a teleoperated mobile robot and Wireless Sensor Networks (WSN) in underground spaces in post-disaster environments. This paper describes about RWSN construction method with using a leaky coaxial cable (LCX) as one means of improving communication connectivity. RWSN with multiple robots, WSN and LCX was constructed to confirm the utilization effect of LCX, and experimental results of communication quality measurement were shown in this paper.

As an information collecting means to grasp the damage situation in an environment where communication is unstable, internal situation such as underground malls and subway premises is hard to confirm when a disaster occurs, we have been studying a system gathering disaster area information by utilizing the rescue robot and Wireless Sensor Network (WSN) built on Mobile Sensor Nodes (MSN). This paper proposes MSN arrangement method to maintain wireless communication connectivity considering a long-time operation of the network. The MSN compares evaluation values of communication quality and remaining energy level with the neighbors and moves to a suitable location maintaining the communication connectivity. The Simulation of change of energy consumption by several communication settings is conducted to confirm the usefulness of the proposed method.

This paper presents a multi-robot communication system for guiding customers and providing information about products and deals in a shop. The purpose of the multi-robot communication is to offer a sense of belonging between the robots and a customer. The robots set in some places cannot move in the store but guide a customer in the direction of what he/she comes for, propagating the robot’s utterance between each of the robots corresponding to the customer’s position and movement. In this study, we conducted the questionnaire investigation and comparative experiments to verify the effect of the proposed approach on customers impressions of the guide robots.

In this study, we propose a flowerpot robot for dementia prevention. The robot consists of a flowerpot, microcomputer and multiple sensors. To interact with a user, the robot expresses its state by posting a comment to a social networking site (SNS) . Additionally, the robot can talk to other nearby flowerpot robots on the SNS. This function encourages mutual approach of the robot users, and is expected to activate communications among the users. In this paper we report on the development situation of hardware.

The system by connecting network a large variety of many robots can respond to various service needs. In this system, RT middleware and robot service network protocols (RSNP) as common basic technologies of robots are used, and the module configuration is adapted for various applications. We have conducted field experiments on community service robots that solve problems. In this case, we developed a system to manage the state of each robot by using RSNP for operation information of multiple kinds of robots and report on the demonstration experiment at Japan robot week 2018.

We have discussed Robot Wireless Sensor Networks (RWSN) construction method for expanding tele-operation area by using mobile robot in post-disaster underground space. In RWSN construction method, mobile robots carry wireless sensor nodes (SNs) and deploy them into the environment. Deployed SNs construct wireless sensor networks by Ad-Hoc connection. In RWSN construction, a wireless communication quality measurement is important to keep connection of communication between the mobile robot and an operator. However, it takes time to construct RWSN utilizing the existing method. Therefore, we developed SNs deployment method utilizing low communication load throughput to propose shortening RWSN construction time. In this paper, we evaluated the proposed method, then it is confirmed that maintaining communication connectivity and shortening RWSN construction time.

Since 2018, we have a project the "Developing the high altitude drone fleet for gathering 3-D mesoscale meteorological information" adopted as a subsidy for Fukushima Innovation Coast Concept Promotion Facility Development. In this research, we develop a telemetry and command communication system for inexpensively manufacturing 920 MHz low power small radio set using LoRa modulation to use our project. In order to construct a drone fleet that performs the same work, we report the definition of the telemetry and the command signals from the project requirement, the actual device specification, and discuss requirement of throughput and multiplexing method from the results of some experiments.

This paper proposes an oblivion model concerning episode memory for communication agents. First, the subjects were instructed to listen to a certain sentence, immediately after which they were asked to freely describe the content. Moreover, one month after this event, the subjects were once again asked to describe the content, and their status of oblivion (quantity, quality) was investigated using the number of characters and idea units. The oblivion model for the communication agent was generated using the results of this investigation.

PEPPER, humanoid robots were designed for entertainment and education purposes. They can especially be employed as an assistance for a teacher or professor. The goal of this research is to create a presentation behavior for Pepper in order to make a more interactive interaction between humans and robots. Choregraphe software and speech recognition available on Pepper robot will be used in order to give presentation and response capabilities when interacting with humans.

Recently, AI equipped robots, smart speakers have appeared in homes, opportunities to come into contact with robots in general households and towns have increased, and various people also have a high possibility of controlling robots. To be able to control the robot smoothly, the robot side needs to understand human intention. In this paper, we propose a system that controls a robot from human ambiguous instructions and images of feet. In this system, Raspberry Pi, USB microphone and USB camera are attached to the mobile robot for control. Control of the robot is performed only with body direction estimated by feet image of the user and voice instructions without attaching special sensors. Ambiguity in speech recognition that do not limit the target using the directive are recognized. Experimental results of 93% success rate when we conducted 10 experiments of here shown the effectiveness of the proposed method.

In recent years, attention has been gaining attention for interchanges between children and the elderly, but in the present situation it is a problem to stay at single-shot event exchanges. Therefore, we developed a system that collaborates with elderly people and children through nurturing avatars in order to carry out continuous daily exchanges, not only for event exchanges but also for verifying the effect. As a result of using this system, the amount of mutual conversation time increased compared to the exchange that has been done conventionally. Therefore, it was suggested that this system has the effect of promoting exchange.

In this paper, we propose a multi-robot interaction system based on audience attention. The multi-robot system consists of multiple robots. Each of the robots has own role for robot interactive theater. The performer robot is programmed to perform a show on the stage; whereas the spectator robot is programmed to behave like an audience. In order to evaluate the immersion in robotic theater, we consider the direction of the user's gaze. Eye contact is important in order to create a life-like communication. Thus, we implemented OpenPose to performer robot for detecting human audiences. During the robot performance in robot theater, spectator robot tends to cheer up the performer robots through active feedbacks. In addition, it further explains the content of the robot performance to human audiences in the theater through verbal speech. We show result what human audiences can focused more to robot theater performance by communicating with spectator robot.

This study aims to develop a plants cultivation system with healing effectiveness by the users perform cultivation works. First, we produced a prototype of the plants cultivation system. Next, we let research participants water it freely for two months. Finally, a questionnaire survey was conducted on the research participants. The result showed that a half of research participants decreased the watering works at the second month than the first month. From the result, it is necessary for the plants cultivation system to have functions of automatically cultivating when the users are satisfied with the cultivating. This paper reports the development status of the plants cultivation system.

This paper describes driving control of low center of gravity two-wheeled mobile robot for tour guide. It is necessary to design the driving control by considering operability and physical contact safety with people. We proposed state feedback control for robot body sway and feedforward control for driving performance. It is possible to estimate driving resistance from body angle during driving. We confirmed to improve driving performance by feedforward control with estimated driving resistance and to ensure physical contact safety with people by effect of state feedback control from experimental results.

The aim of our research is constructing a method which allows robot to learn and generate the behavior that an individual prefers. In our method, observed data (including facial expressions) are used to generate and update the model which estimates/predicts change of mental state during human-robot interaction. In an user study to verify the proposed method, the effect of the un-expected factor which changes a mental state/emotion must be decreased. For this reason, we developed whack-a-mole game system as the situation which makes him/her concentrate to and reacts quickly.

Japanese farmers have steadily decreased and aged in recent years. Hence, an autonomous agricultural robot is expected to overcome these problems. In this paper, a robot forklift is proposed for automated container handling. The robot forklift has GNSS receivers, a high-resolution camera, and a 3D-LiDAR. Our proposed method can detect containers and estimate a distance and horizontal angles between the forklift and its surface by Deep Learning and 3D-LiDAR. Many previous studies have been investigated to detect the objects by image segmentation based on color and geometric characteristics of the target. However, these studies have the predicament that has a weakness to the noise of colors. The experimental results showed that it successfully worked under the real environment.

Raspberries contain ingredients with health benefits, and mechanizing the harvesting work will result in revenue. The purpose of this research is to classify fruits by maturity by machine learning based on reflection spectrum. In the visible reflection spectroscopy, since spectra in which all components contained in the fruit skin etc. overlap are measured, it is impossible to estimate all the individual components in the fruit. In this research, in order to solve this problem, it is necessary to trace the reflection spectrum of individuals of fruits for a certain period of time, utilize machine learning, and finely classify fruits by maturity derived from changes in components in fruits did.

In this research we aim to develop the water drone equipped with the lamp to induce duck repelling behavior. In this report, we will consider an automatic traveling system using an infrared camera and an infrared laser sensor for detecting the duck at night and driving the lotus field automatically.

In automatic harvesting of vegetables, since crops does not grow in a straight line, it is necessary to track a harvester to a path corresponds to the position of farm products. On the other hand, tracked vehicles with on/off drive are often used for vegetable harvesters, and the vehicles have constraints such as the minimum turning radius. Therefore, the curvature radius of a generated path becomes smaller than the minimum turning radius in some cases when the paths are generated with spline interpolation etc. Hence, we propose a path generation method considering constraints of vehicle characteristics using Model Predictive Control, and we shows its effectiveness by numerical simulations.

Currently, there is a serious safety issue for mowing machines such as brush cutter. Therefore, the authors have researched and developed a remote controlled compact mowing robot TMR-R11Ver.2. In this paper, using the proposed speed control method, the work efficiency of mowing robot is evaluated. In the proposed speed control method, load torque of mowing motor is used for drive motor control. We modeled the developed mowing robot and analyzed the relationship between load torque, travel speed, consumption current and operating time during grass cutting by simulation. Effect of the grass density is evaluated. The results show the high success rate and short operating time under present simulation conditions.

As a herbicide-free weeding method, our research group are developing an autonomous running paddy weeding robot that can travel by a simple operation. The paddy weeding robot removes the weeds by churning up the soil and inhibits the growth of the weeds by blocking-off sunlight. In this paper, we propose a turning control method that performs a turning motion for repeating autonomous running. The proposed turning controller consists of three functions, final rice plant detection，turning control, and rows of rice plant detection. The experiment results demonstrate that the proposed turning control method is effective in turning motion for the paddy weeding robot.

In Japanese mountainous region, there are many levee slopes, in which weeds are grown and effective vegetation are needed. The string trimmers are used to conserve the vegetation of levee slopes. At the same time, however, it is true that some of the users, especially of the elders, have an accident by them. The Japanese government, accordingly, has been promoting the automation of agricultural machine like weed cutter robots, to prevent accidents.

In the development of weed cutter robots for levee slope, sideslip is an important issue. We confirmed the effectiveness of camber angle for the suppression of the sideslip. However, sufficient consideration has not been made on the camber angle suitable for sideslip suppression. In this paper, the author discussed the relationship between wheel differences and behavioral differences of vehicle with camber angle.

An agricultural robot is highly required due to decreasing of agricultural working population. So far, fully automatic agricultural robots without any manual operations have been proposed. However, there are still many problems to be solved regarding economical and technical side of putting this robot during the harvesting season in practical use. To cope with the problem, in this paper, we propose a tomato harvesting and deleafing assist robot system that an operator can select tomatoes to be harvested and deleafed. In the proposed system, based on the harvesting target selection system by operator, we will construct a deleafing system by expanding the selection method.