Details of a Researcher - Takahashi, Masaki

Papers - Takahashi, Masaki

Division display 61 - 80 of about 201 ／ All the affair displays >>

Model predictive control of autonomous drone considering model of birds aimed at inducing a flock of birds

Hamabe S., Takahashi M.

AIAA Scitech 2020 Forum (AIAA Scitech 2020 Forum) 1 PartF 2020

ISSN 9781624105951

© 2020, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. In recent years, bird strikes are increasing with the increase in the number of airliner flights. Therefore, measures against bird strikes are required. Collisions between an aircraft and a flock of birds during takeoff and landing are numerous and can cause serious damage. Related works have been conducted to prevent entry of a flock of birds into the runway. The purpose of this study was to guide a flock of birds into avoiding the runway using an autonomous drone. Guidance was successful in simulations and experiments, provided that the drone was located near a flock of birds. Therefore, in this study, assuming that the drones are placed around the runway and far from the flock, we propose a control method to move the drone towards the flock and guide the birds to prevent bird strikes. In order to realize interception and guidance under such conditions, we predict the trajectories of the bird flock using a model of birds. Setting the target position of the drone on the predicted trajectories allows the drone to intercept the birds without unnecessary movement of the drone. Because the drone is in the same speed range as the birds, we determine the drone’s control input so as to satisfy the velocity constraint using model predictive control (MPC). In this study, we propose a control method for autonomous drones that considers the model of birds, aimed at guiding a flock of birds to prevent bird strike. We verify the performance of the proposed method by conducting numerical analysis based on a model of a realistic runway.

Access to Document (DOI)

A Multimodal Path Planning Approach to Human Robot Interaction Based on Integrating Action Modeling

Kawasaki Y., Yorozu A., Takahashi M., Pagello E.

Journal of Intelligent and Robotic Systems: Theory and Applications (Journal of Intelligent and Robotic Systems: Theory and Applications) 100 ( 3-4 ) 955 - 972 2020

ISSN 09210296

© 2020, Springer Nature B.V. To complete a task consisting of a series of actions that involve human-robot interaction, it is necessary to plan a motion that considers each action individually as well as in relation to the following action. We then focus on the specific action of “approaching a group of people” in order to accurately obtain human data that is used to make the performance of tasks involving interactions with multiple people more smooth. The movement depends on the characteristics of the important sensors used for the task and on the placement of people at and around the destination. Considering the multiple tasks and placement of people, the pre-calculation of the destinations and paths is difficult. This paper thus presents a system of navigation that can accurately obtain human data based on sensor characteristics, task content, and real-time sensor data for processes involving human-robot interaction (HRI); this method does not navigate specifically toward a previously determined static point. Our goal was achieved by using a multimodal path planning based on integration of action modeling by considering both voice and image sensing of interacting people as well as obstacle avoidance. We experimentally verified our method by using a robot in a coffee shop environment.

Access to Document (DOI)

Terrain relative navigation enhanced with sar for moon’s shadowed regions

Hidaka M., Takahashi M., Ishida T., Fukuda S.

AIAA Scitech 2020 Forum (AIAA Scitech 2020 Forum) 1 PartF 2020

ISSN 9781624105951

This study proposes a terrain relative navigation (TRN) method to estimate spacecraft positions in the shadowed regions of the moon with high precision by combining the camera image and the synthetic aperture radar (SAR) data. TRN, which estimates the position of a spacecraft by comparing the preliminary terrain information stored in a database with the observed terrain, is an effective method of correcting the drift errors of inertial measurement units (IMUs). Optical cameras are primarily used for TRN sensors to observe the terrain; however, shadow images do not have enough characteristic points to match with the database. SAR is a form of radar that can generate high-resolution photo-like images, regardless of light conditions. In this study, we propose the use of SAR as navigation sensors and demonstrate effectiveness of SAR.

Access to Document (DOI)

Pilot-centered evaluation of flight-deck interval management control laws using an a320 simulator

Riedel T., Takahashi M., Itoh E., Frost P., Feuerle T.

Journal of Aircraft (Journal of Aircraft) 57 ( 5 ) 974 - 984 2020

ISSN 00218669

Recent research on flight-deck interval management technology, including the findings of a flight-test demonstration conducted by NASA, led to the development of an alternative control logic for flight-deck interval management, termed as “interval management–speed planning.” As this logic demonstrated promising results in numerical simulations, for the first time, test pilots and airline pilots were invited to evaluate the system in an Airbus A320 simulator, using an avionics setup similar to the one employed in NASA’s flight test, on an approach for the Tokyo International Airport. A qualitative and quantitative survey was conducted in which the pilots’ opinions pertaining to their acceptance of the interval management operation and comprehension of the system behavior using both the original and alternative control logic, as well as a modified graphical user interface, were sought. The results indicated that the new control logic was well accepted by the pilots, confirming some of the results and hypotheses from the numerical simulation; nevertheless, it was indicated that further research on the user interface is needed to make full use of the logic’s capability.

Access to Document (DOI)

Optimisation of interval management - Speed planning using SMPSO

Riedel T., Takahashi M., Itoh E.

Aeronautical Journal (Aeronautical Journal) 124 ( 1281 ) 1819 - 1845 2020

ISSN 00019240

Recent research on Flight-deck Interval Management (FIM), a modern technology for increasing safety and improving airspace and runway utilisation through self-spacing, has led to the development of a new rule-based logic for FIM, namely Interval Management - Speed Planning (IM-SP). In an initial benchmark study, IM-SP showed good spacing performance with a significant reduction in speed commands, a major area of concern with previous FIM logics, resulting in a lower burden on the flight crew during FIM operation. Nevertheless, there remains scope for improvement in other aspects, such as fuel burn. In this study, the internal cost function of IM-SP is further analysed and optimised using speed-constrained multi-objective particle swarm optimisation to improve the performance of IM-SP under the multiple objectives of FIM. The optimisation renders new settings that address the problem areas, improve the speed commands and enhance the overall quality of IM-SP. Two distinctive solutions, viz. a spacing performance optimised setting and a fuel burn optimised setting, are further analysed and discussed, and directions for follow-up research are explored.

Access to Document (DOI)

Reinforcement Learning of Trajectory Distributions: Applications in Assisted Teleoperation and Motion Planning

Ewerton M., Maeda G., Koert D., Kolev Z., Takahashi M., Peters J.

IEEE International Conference on Intelligent Robots and Systems (IEEE International Conference on Intelligent Robots and Systems) 4294 - 4300 2019.11

ISSN 9781728140049

© 2019 IEEE. The majority of learning from demonstration approaches do not address suboptimal demonstrations or cases when drastic changes in the environment occur after the demonstrations were made. For example, in real teleoperation tasks, the demonstrations provided by the user are often suboptimal due to interface and hardware limitations. In tasks involving co-manipulation and manipulation planning, the environment often changes due to unexpected obstacles rendering previous demonstrations invalid. This paper presents a reinforcement learning algorithm that exploits the use of relevance functions to tackle such problems. This paper introduces the Pearson correlation as a measure of the relevance of policy parameters in regards to each of the components of the cost function to be optimized. The method is demonstrated in a static environment where the quality of the teleoperation is compromised by the visual interface (operating a robot in a three-dimensional task by using a simple 2D monitor). Afterward, we tested the method on a dynamic environment using a real 7-DoF robot arm where distributions are computed online via Gaussian Process regression.

Access to Document (DOI)

Learning Trajectory Distributions for Assisted Teleoperation and Path Planning

Ewerton M., Arenz O., Maeda G., Koert D., Kolev Z., Takahashi M., Peters J.

Frontiers in Robotics and AI (Frontiers in Robotics and AI) 6 2019.09

Several approaches have been proposed to assist humans in co-manipulation and teleoperation tasks given demonstrated trajectories. However, these approaches are not applicable when the demonstrations are suboptimal or when the generalization capabilities of the learned models cannot cope with the changes in the environment. Nevertheless, in real co-manipulation and teleoperation tasks, the original demonstrations will often be suboptimal and a learning system must be able to cope with new situations. This paper presents a reinforcement learning algorithm that can be applied to such problems. The proposed algorithm is initialized with a probability distribution of demonstrated trajectories and is based on the concept of relevance functions. We show in this paper how the relevance of trajectory parameters to optimization objectives is connected with the concept of Pearson correlation. First, we demonstrate the efficacy of our algorithm by addressing the assisted teleoperation of an object in a static virtual environment. Afterward, we extend this algorithm to deal with dynamic environments by utilizing Gaussian Process regression. The full framework is applied to make a point particle and a 7-DoF robot arm autonomously adapt their movements to changes in the environment as well as to assist the teleoperation of a 7-DoF robot arm in a dynamic environment.

Access to Document (DOI)

Stair climbing ability in patients with early knee osteoarthritis: Defining the clinical hallmarks of early disease

Iijima H., Eguchi R., Shimoura K., Aoyama T., Takahashi M.

Gait and Posture (Gait and Posture) 72 148 - 153 2019.07

ISSN 09666362

© 2019 Elsevier B.V. Background: A growing clinical interest has been shown towards identifying knee osteoarthritis (OA) patients at earlier stages. The early detection of knee OA may allow for more effective interventions. Research question: The aim of this study was to determine the discriminative ability of a stair-climb test (SCT) in identifying patients with early knee OA, and to determine if descending stair time during the SCT is better than ascending stair time for the identification of these patients. Methods: This study was a secondary, cross-sectional analysis of baseline data from a randomized controlled trial. Adults with moderate to severe knee pain were enrolled (n = 57; mean age 58.9 years; 71.9% women). Each participant performed an 11-step SCT (11-SCT) while wearing shoes with a pressure sensor insole. A receiver operating characteristic analysis was used to examine the discriminative power of 11-SCT for identifying early knee OA (Kellgren and Lawrence grade 1). The discriminative power was also compared between the ascending and descending 11-SCT time as evaluated by the pressure sensor. Results: The 11-SCT time in patients with early knee OA was 0.55 s longer than that in those with symptomatic non-radiographic OA. A one-second increase in the 11-SCT time was significantly associated with 1.9-fold increased odds of early knee OA being present. The 11-SCT value with the best balance of sensitivity and specificity for identifying early knee OA was 8.33 s (area under the curve: 0.711). The descending time was not significantly better than the ascending time for identifying early knee OA. Significance: This study determined the time values of an 11-SCT that may be useful for identifying early knee OA patients. These preliminary findings may serve as the foundation for future studies investigating the clinical hallmarks associated with early knee OA.

Access to Document (DOI)

Effects of interaction between varus thrust and ambulatory physical activity on knee pain in individuals with knee osteoarthritis: an exploratory study with 12-month follow-up

Iijima H., Aoyama T., Eguchi R., Takahashi M., Matsuda S.

Clinical Rheumatology (Clinical Rheumatology) 38 ( 6 ) 1721 - 1729 2019.06

ISSN 07703198

© 2019, International League of Associations for Rheumatology (ILAR). Introduction: This study aimed to examine the interaction effect between ambulatory physical activity (PA) and varus thrust on knee pain in individuals with knee osteoarthritis (OA). Method: Subjects (n = 207; mean age: 73.1 years, 71.5% women) in orthopedic clinics with diagnosed knee OA (Kellgren/Lawrence grade ≥ 1) were enrolled in this 12-month observational cohort study. Participants underwent gait observation for varus thrust assessment and pedometer-based ambulatory PA measurements at baseline and 12-month follow-up. Knee pain intensity was assessed using the Japanese Knee Osteoarthritis Measure pain subscale as a primary outcome measure. Multiple linear regression analyses were performed to evaluate ambulatory PA–thrust interaction on knee pain intensity. Results: Ninety-two subjects (mean age, 73.4 years; 68.5% women) completed the 12-month follow-up assessment. Baseline ambulatory PA–thrust interaction was significant (P = 0.017) in the cross-sectional analysis, adjusting for covariates, which yielded R2 = 0.310. Subgroup analysis showed that varus thrust was significantly associated with worse knee pain in subjects walking ≥ 5000 steps/day adjusting for covariates (beta: 7.94; 95% CI: 3.82, 12.1; P < 0.001) with a higher predictive ability (R2 = 0.664). In contrast, ambulatory PA–thrust interaction in the longitudinal analysis showed no significant association with knee pain changes. Conclusions: Ambulatory PA interacted with varus thrust in the association with knee pain, as coexisting high ambulatory PA and varus thrust had the strongest association with higher knee pain. Maximal pain relief effects might be achieved when both ambulatory PA and varus thrust are treated simultaneously, rather than treating each separately.

Access to Document (DOI)

Design of Vibration-Isolating Bed for Ambulances Using Inerter

Koyanagi R., Takahashi M.

2019 12th Asian Control Conference, ASCC 2019 (2019 12th Asian Control Conference, ASCC 2019) 567 - 572 2019.06

ISSN 9784888983006

© 2019 JSME. A vibration-isolating bed for ambulances using an inerter is proposed. The structure of the bed is assumed to consist of a bottom frame fixed to the ambulance floor and an upper frame on which the stretcher is placed. Both are connected only by the suspension, including the inerter, in the vertical and horizontal directions. Because the inerter can reduce the gain of the high frequency band and delay the response, the time when maximum acceleration is applied to bed can be shortened by passively in a limited space in ambulance. We consider the bed as a plant and the suspension as a controller, and then formulate the state equation. To reduce the acceleration within the movable range of the bed, the controller that minimizes the H∞ norm of the transfer function from the external force acting on the bed to acceleration is calculated by bilinear matrix inequality. The transfer function determines the structure and coefficient of the suspension. The usefulness of the bed is verified by numerical simulations.

Walking Measurement System Including Turning Motion Assessment Using Depth Sensors

Ono T., Takahashi M.

2019 12th Asian Control Conference, ASCC 2019 (2019 12th Asian Control Conference, ASCC 2019) 861 - 866 2019.06

ISSN 9784888983006

© 2019 JSME. Recently, with the aging of the population, fall accidents of elderly people are increasing. In order to reduce fall accidents, it is necessary to detect the walking ability decline of elderly people. People with medical knowledge, such as physiotherapists, qualitatively evaluate and judge walking ability from experiences and tendencies by a walking test. However, in this case, it is difficult to examine many subjects and to evaluate them quantitatively. Therefore, it is increasing to the demand for walking measurement systems using instruments. Requirement of this system is not that it can be measured with high accuracy but that it is simple, inexpensive and accurate enough for gait ability evaluation. Therefore, three-dimensional motion analysis device that are high accuracy but expensive and need to be equipped with markers are not suitable.

Satellite Attitude Control and Power Tracking with VSCMGs during Large-angle and Agile Attitude Maneuvers

Yoshihara H., Takahashi M.

2019 12th Asian Control Conference, ASCC 2019 (2019 12th Asian Control Conference, ASCC 2019) 1613 - 1618 2019.06

ISSN 9784888983006

© 2019 JSME. In recent years, the use of VSCMG/IPACS for satellites has been considered, because Earth observation satellites are required to have agility and miniaturization. However, in previous research, the control laws of VSCMG/IPACS at large-angle and agile attitude maneuvers have not been studied. In this study, we clarify that it is impossible for conventional methods to achieve large-angle and agile attitude maneuvers, while accurately tracking power. Therefore, we propose a system capable of simultaneous satellite attitude control and power tracking with VSCMG during large-angle and agile attitude maneuvers. The proposed steering law switches the steering laws according to singularities, and enables the satellite to change attitude quickly without power errors. In addition, numerical simulations demonstrate through comparison that the proposed steering law is more useful than conventional steering laws.

Insole-Based Estimation of Vertical Ground Reaction Force Using One-Step Learning with Probabilistic Regression and Data Augmentation

Eguchi R., Takahashi M.

IEEE Transactions on Neural Systems and Rehabilitation Engineering (IEEE Transactions on Neural Systems and Rehabilitation Engineering) 27 ( 6 ) 1217 - 1225 2019.06

ISSN 15344320

© 2001-2011 IEEE. An insole-based estimation of the vertical ground reaction force (vGRF) is proposed as an alternative to costly force plates for the evaluation of pathological gait. However, machine learning techniques for estimation still rely on the use of force plates. Moreover, measuring plural walking steps in order to prevent overfitting induces fall risks and physically taxes the patients. Therefore, this paper presents an accessible and efficient learning scheme for the insole-based estimation of vGRF. In this system, we employ a low-cost scale as an alternative to force plates. Then, we use Gaussian process regression (GPR) to learn a model in order to estimate vGRF without overfitting of small-sized data sets corrupted by measurement errors and noise of the devices. In addition, we propose a 'one-step learning' scheme based on a probabilistic data augmentation. This approach augments actual measurements of a minimum (just one) walking step to a virtual data set for plural steps by considering their typical variability between steps. In experiments, the GPR models learned from two walking steps estimated vGRF with mean errors of 8% or under for entire/local magnitudes. Moreover, the learning from one step with probabilistic augmentation enhanced the estimation accuracy.

Access to Document (DOI)

Spatiotemporal and kinetic gait analysis system based on multisensor fusion of laser range sensor and instrumented insoles

Eguchi R., Yorozu A., Takahashi M.

Proceedings - IEEE International Conference on Robotics and Automation (Proceedings - IEEE International Conference on Robotics and Automation) 2019-May 4876 - 4881 2019.05

ISSN 9781538660263

© 2019 IEEE. Tracking of human legs during walking are key technologies for gait analysis evaluating the movement function of the elderly and patients with gait disorders. Although the motion capture cameras are the gold standard method for gait analysis because of their high accuracy, they are not always accessible in clinical sites because of their cost, scale, and usability. In response, a laser range sensor (LRS), which is used for obstacle avoidance and human detection of mobile robots, has recently been employed for tracking of leg motions. Some previous studies set LRS at shin height and tracked leg motions during walking using three or five observation patterns and the Kalman filtering and data association methods. However, these systems had difficulty in tracking during walking along a circular trajectory including frequent overlaps and occlusions of legs. Therefore, this paper presents a spatiotemporal and kinetic gait analysis system using a single LRS and instrumented insoles and proposes a multisensor fusion algorithm for tracking leg motions. The instrumented insoles are in-shoe devices embedded force sensors and can detect accurate timings of gait events via force sensing. The system identifies gait phases by the fusion algorithm and switches acceleration input added to motion models of tracked legs for the Kalman filter and data association. The tracking performance of the proposed system was evaluated by measuring walking on a circular trajectory in experiments.

Access to Document (DOI)

Proximal gait adaptations in individuals with knee osteoarthritis: A systematic review and meta-analysis

Iijima H., Shimoura K., Ono T., Aoyama T., Takahashi M.

Journal of Biomechanics (Journal of Biomechanics) 87 127 - 141 2019.04

ISSN 00219290

© 2019 Elsevier Ltd Clarifying proximal gait adaptations as a strategy to reduce knee joint loading and pain for individuals with knee osteoarthritis (OA) contributes to understanding the pathogenesis of multi-articular OA changes and musculoskeletal pain in other joints. We aimed to determine whether biomechanical alterations in knee OA patients during level walking is increased upper trunk lean in the frontal and sagittal planes, and subsequent alteration in external hip adduction moment (EHAM) and external hip flexion moment (EHFM). A literature search was conducted in PubMed, PEDro, CINAHL, and Cochrane CENTRAL through May 2018. Where possible, data were combined into a meta-analysis; pooled standardized mean differences (SMD) of between knee OA patients and healthy adults were calculated using a random-effect model. In total, 32 articles (2037 participants, mean age, 63.0 years) met inclusion criteria. Individuals with knee OA had significantly increased lateral trunk lean toward the ipsilateral limb (pooled SMD: 1.18; 95% CI: 0.59, 1.77) along with significantly decreased EHAM. These subjects also displayed a non-significantly increased trunk/pelvic flexion angle and EHFM. The GRADE approach judged all measures as “very low.” These results may indicate that biomechanical alterations accompanying knee OA are associated with increased lateral trunk lean and ensuing alterations in EHAM. Biomechanical alterations in the sagittal plane were not evident. Biomechanical adaptations might have negative sequelae, such as secondary hip abductor muscle weakness and low back pain. Thus, investigations of negative sequelae due to proximal gait adaptations are warranted.

Access to Document (DOI)

Concurrent validity and measurement error of stair climb test in people with pre-radiographic to mild knee osteoarthritis

Iijima H., Shimoura K., Eguchi R., Aoyama T., Takahashi M.

Gait and Posture (Gait and Posture) 68 335 - 339 2019.02

ISSN 09666362

© 2018 Elsevier B.V. Background: Stair climbing is the task first affected in patients with knee osteoarthritis (OA); therefore, the precise measurement of time required to climb stairs is important to identify mobility limitations, particularly in the early phase of knee OA. Research question: This study aimed to examine the test-retest reliability, measurement error, and concurrent validity of the stopwatch-based stair-climb test (SCT) in adults with pre-radiographic to mild knee OA. Methods: Fifty-nine participants (mean age, 59.1 [range, 50–69] years; 72.9% female) with Kellgren and Lawrence grade ≤2 disease underwent an 11-step SCT (11-SCT) in accordance with the Osteoarthritis Research Society International recommended method while wearing pressure sensor-mounted standard shoes that is used as a gold standard procedure. Test-retest reliability, measurement errors, and the concurrent validity of the stopwatch-based 11-SCT were evaluated. Results: The test-retest reliability of the stopwatch-based 11-SCT was excellent (intra-class correlation coefficient 1,1 [ICC 1,1 ], 0.952; 95% confidence interval [CI], 0.560 to 0.985; p < 0.001) and the minimal detectable change 95 was 0.102 s. Concurrent validity was excellent (ICC 2,1 : 0.957; 95% CI: 0.661 to 0.986; p < 0.001). Significance: The stopwatch-based 11-SCT had high test-retest reliability and high concurrent validity, which justify its clinical use for identifying mobility limitations in individuals with pre-radiographic to mild knee OA. A difference of 0.2 s in the stopwatch-based 11-SCT time would be considered a true difference beyond a 95% measurement error.

Access to Document (DOI)

Trunk movement asymmetry associated with pain, disability, and quadriceps strength asymmetry in individuals with knee osteoarthritis: a cross-sectional study

Iijima H., Eguchi R., Aoyama T., Takahashi M.

Osteoarthritis and Cartilage (Osteoarthritis and Cartilage) 27 ( 2 ) 248 - 256 2019.02

ISSN 10634584

© 2018 Osteoarthritis Research Society International Objective: This study examined 1) the clinical relevance of trunk movement asymmetry, which was evaluated using a trunk-mounted inertial measurement unit (IMU), and 2) the relationship between trunk movement asymmetry and lower limb muscle strength asymmetry in individuals with knee osteoarthritis (OA). Design: One-hundred-thirty-one participants (mean age, 74.2 years; 71.8% female; Kellgren and Lawrence [K&L] grade ≥1) underwent gait analysis at their preferred pace for IMU-based measurement of trunk movement asymmetry (harmonic ratio [HR] and improved HR). The isometric strength of quadriceps and hip abductors was evaluated using a hand-held dynamometer. Pain and disability level were evaluated using a validated self-reported questionnaire. Multiple regression analyses with covariate adjustment were performed to examine the relationship between trunk movement asymmetry (independent variable) and pain, disability level, or muscle strength asymmetry (dependent variables). Results: Individuals with severe knee OA (K&L grade ≥3) had increased trunk movement asymmetry in the medio-lateral axis compared to those with a K&L grade of 1. Increased trunk movement asymmetry was associated with a greater knee pain and disability. The increased trunk movement asymmetry was significantly associated with an increase in the asymmetry of quadriceps strength, but not with asymmetry in the strength of hip abductor. Conclusion: Our findings indicate that increased medio-lateral trunk movement asymmetry may be an indicator of impairment, rather than adaptation, in individuals with knee OA. This preliminary finding warrants validation by future study. Paying close attention to medio-lateral trunk movement asymmetry may be key to our understanding of OA-related pain and disability.

Access to Document (DOI)

Evaluating applied flight-deck interval management using monte carlo simulations on the k-supercomputer

Riedel T., Takahashi M., Tatsukawa T., Itoh E.

Transactions of the Japan Society for Aeronautical and Space Sciences (Transactions of the Japan Society for Aeronautical and Space Sciences) 62 ( 6 ) 299 - 309 2019

ISSN 05493811

© 2019 The Japan Society for Aeronautical and Space Sciences. Increasing air traffic and airport congestion calls for solutions to improve traffic throughput. In this paper, we evaluate one solution, Flight-deck Interval Management and its underlying logic, particularly in terms of its utility. An environment to simulate traffic engaged in interval management on the most common arrival routes toward Tokyo International Airport was set up and run as a large-scale Monte Carlo simulation on the K-Supercomputer. The results show that under low wind conditions in particular, spacing goal times with a standard deviation of less than 5 s can be achieved and a speed tolerance mode can reduce the number of speed commands without compromising performance.

Access to Document (DOI)

Signal processing of landing radar considering irradiated surface characteristics using convolutional neural networks

Hidaka M., Takahashi M., Ishida T., Kariya K., Mizuno T., Fukuda S.

AIAA Scitech 2019 Forum (AIAA Scitech 2019 Forum) 2019

ISSN 9781624105784

© 2019, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. In this paper, a signal-processing method for a lunar lander using deep learning is proposed. The ability for pinpoint soft landing on a lunar/planetary surface broadens the range of scientific and exploration missions. To perform pinpoint landing, measurement of the relative velocity with respect to the surface is essential. Landing radar is a sensor that measures the relative velocity. To measure the velocity, the landing radar irradiates the surface with a pulse wave and observes the Doppler shift. High-precision measurement on complex terrains, a crater, or a slope has always been the problem of landing radar because the irradiated terrains strongly affect the accuracy. We propose a measurement system that performs with high accuracy on complex terrains using convolutional neural networks. Moreover, we confirm that the proposed method could improve the measurement accuracy compared with the existing method.

Access to Document (DOI)

Service robot using estimation of body direction based on gait for human-robot interaction

Yorozu A., Takahashi M.

Advances in Intelligent Systems and Computing (Advances in Intelligent Systems and Computing) 867 199 - 209 2019

ISSN 9783030013691

© 2019, Springer Nature Switzerland AG. Recently, there have been several studies on the research and development of service robots, and experimental results in real environments have been reported. To realize socially acceptable human-robot interaction for service robots, human recognition, including not only position but also body direction, around the robot is important. Using an RGB-D camera, it is possible to detect the posture of a person. However, because the viewing angle of the camera is narrow, it is difficult to recognize the environment around the robot with a single device. This study proposes the estimation of the body direction based on the gait, that is, not only the position and velocity, but also the state of the legs (stance or swing phase), using laser range sensors installed at shin height. We verify the effectiveness of the proposed method for several patterns of movement, which are seen when a person interacts with the service robot.

Access to Document (DOI)

　 Previous page - Next page 　