Details of a Researcher - Takahashi, Masaki

Papers - Takahashi, Masaki

Division display 1 - 20 of about 201 ／ All the affair displays >>

Estimation of Three-Dimensional Ground Reaction Forces During Walking and Turning Using Insole Pressure Sensors Based on Gait Pattern Recognition

Eguchi R., Takahashi M.

IEEE Sensors Journal (IEEE Sensors Journal) 23 ( 24 ) 31278 - 31286 2023.12

ISSN 1530437X

Three-dimensional ground reaction forces (3D GRFs) in various gait patterns (e.g., walking and turning) provide essential information for clinical assessment. Previous scholars proposed to estimate GRFs from insoles with few pressure sensors using machine learning techniques. However, estimating GRFs during turning using a model learned from walking proved difficult because relationships between plantar pressure and GRFs vary according to the gait pattern. In this study, GRFs were estimated based on gait pattern recognition. GRF estimation models were learned from various gait patterns in advance. These models represented relationships between the insole measurements and GRFs using a Gaussian process regression (GPR). The GRFs were estimated from the insole measurements using a maximum likelihood (ML) model recognized as the current gait pattern. The ML model had the largest time frames, in which its standard deviation (SD) of the probabilistically estimated vertical GRF was the smallest among all the models (i.e., its training data were closest to the inputs), in the initial and terminal stance phases (subphases). In addition, the time frames of the contralateral leg were considered to enhance the recognition accuracy based on interlimb coordination. Experiment results showed that the proposed system, which used a common model learned from all the gait patterns for the vertical GRF and the ML model in the subphases of both legs for the horizontal GRFs, estimated 3D GRFs during walking and turning with different curvatures and directions with higher accuracy. The system can be widely applied for clinical walking tests and monitoring in daily life.

Access to Document (DOI)

Gait Measurement System Utilizing Missing Point Clouds Caused by Feet With Apple LiDAR Camera

Takahashi M., Takahashi M.

IEEE Sensors Letters (IEEE Sensors Letters) 7 ( 10 ) 2023.10

Gait measurement in clinical settings is often limited to qualitative evaluations despite the availability of various measurement devices. Although marker-less motion capture techniques, such as Kinect, have been attracting attention recently, they have rarely been introduced in the field because of the complexity of measurement operations. Therefore, in this letter, we propose a gait measurement algorithm that uses the light detection and ranging (LiDAR) camera of an iPad, a widely used mobile device. First, as part of the floor was hidden by the walking foot, the missing points in the floor point cloud were detected as feet. Subsequently, we placed the potentials at the detected foot positions, summed them over a series of frames, and obtained the local maximum value to obtain the ground position during walking. Spatial gait parameters, such as step length and width, were obtained from the ground position. In an accuracy verification experiment with 13 elderly subjects, the average absolute error of the step length was 28 mm, and that of the step width was 19 mm. These results indicated that the values estimated using the proposed method are useful for evaluating clinical tasks.

Access to Document (DOI)

Simultaneous Measurement of Respiratory and Pulse Information Using a Patch-Type Sensor for Sleep Apnea Hypopnea Detection

Ota Y., Yoshida T., Nagaoka H., Takahashi M.

IEEE Sensors Letters (IEEE Sensors Letters) 7 ( 10 ) 2023.10

Sleep apnea hypopnea syndrome (SAHS) is one of the most common sleeping disorders. However, numerous people suffering from SAHS are left undiagnosed. SAHS causes arousal reactions, sleep fragmentation, and hypoxemia, thus leading to excessive daytime sleepiness, lack of concentration, and a high risk of cardiovascular complications. Although polysomnography is the standard way of SAHS screening, its high cost, limited accessibility, and requirement of operational skills hinder its applicability for SAHS diagnoses. Therefore, this letter proposes a novel approach for SAHS detection using a patch-type photoplethysmography sensor. As an initial study, both respiratory and pulse information are simultaneously measured, and the trend of pulse variation, which reportedly has a strong correlation to oxygen saturation (SpO2), is thereby clarified. In addition, more real-time measurement of the pulse rate was possible on the chest compared to the fingertip. SAHS detection by incorporating respiratory and pulse information is beneficial for complete and credible screening and risk assessment of cardiovascular complications with less invasive devices. Therefore, this simple patch-type sensor can replace invasive, expensive, and limiting screening methods to facilitate the diagnosis of SAHS.

Access to Document (DOI)

The modeling and analysis for the natural frequency adjustments of Vehicle Manipulator for the ITER Blanket Remote Handling System

Tomoyuki I., Yuto N., Nobukazu T., Masaki T.

Fusion Engineering and Design (Fusion Engineering and Design) 192 2023.07

ISSN 09203796

This paper describes a method to adjust the natural frequency of the Vehicle Manipulator of the ITER Blanket Remote Handling System (BRHS). The natural frequency was adjusted to avoid the seismic resonance by using the redundancy of degree of freedom of the Vehicle Manipulator. The Vehicle Manipulator was modeled and analyzed with lumped mass model with joint stiffness. The analysis has shown that the natural frequencies of the Vehicle Manipulator can be changed from 0.5 Hz to 12 Hz in the gripping posture of the First Wall in the vacuum vessel, suggesting that the dominant seismic frequency around 7Hz-9 Hz can be avoided.

Access to Document (DOI)

Goal Inference via Corrective Path Demonstration for Human-Robot Collaboration

Ohnishi F., Kawasaki Y., Takahashi M.

Lecture Notes in Networks and Systems (Lecture Notes in Networks and Systems) 577 LNNS 15 - 28 2023

ISSN 23673370

Recently, collaborative robots, such as collaborative delivery robots, have been expected to improve the work efficiency of users. For natural human-robot collaboration, it is necessary to infer the appropriate goal position to transport instruments, where the user’s convenience and the surrounding environment are considered. In conventional research, the goal is inferred by demonstrating the user’s desired positions, but position demonstration requires many trials to obtain the inference model, which is burdensome for the user. Therefore, we focus on the user’s correction of the robot position and generate multiple position samples from the user’s corrective path. In addition, these position samples are weighted based on the implicit intention of the correction to learn both the desired and undesired positions. Consequently, the robot improves goal inference in fewer trials. The effectiveness of the proposed method was evaluated by experiment that simulated human-robot collaborative environments.

Access to Document (DOI)

Switching Head-Tail Funnel UNITER for Dual Referring Expression Comprehension with Fetch-and-Carry Tasks

Korekata R., Kambara M., Yoshida Y., Ishikawa S., Kawasaki Y., Takahashi M., Sugiura K.

IEEE International Conference on Intelligent Robots and Systems (IEEE International Conference on Intelligent Robots and Systems) 3865 - 3872 2023

ISSN 21530858

This paper describes a domestic service robot (DSR) that fetches everyday objects and carries them to specified destinations according to free-form natural language instructions. Given an instruction such as 'Move the bottle on the left side of the plate to the empty chair,' the DSR is expected to identify the bottle and the chair from multiple candidates in the environment and carry the target object to the destination. Most of the existing multimodal language understanding methods are impractical in terms of computational complexity because they require inferences for all combinations of target object candidates and destination candidates. We propose Switching Head-Tail Funnel UNITER, which solves the task by predicting the target object and the destination individually using a single model. Our method is validated on a dataset based on a standard dataset for Vision-and-Language Navigation with object manipulation tasks. The results show that our method outperforms the baseline method in terms of language comprehension accuracy. Furthermore, we conduct physical experiments in which a DSR delivers standardized everyday objects in a standardized domestic environment as requested by instructions with referring expressions. The experimental results show that the object grasping and placing actions are achieved with success rates of more than 90 %.

Access to Document (DOI)

Rush-Out Risk Mapping from Human Operational Commands Considering Field Context<sup>∗</sup>

Ohnishi F., Kawasaki Y., Takahashi M.

IEEE International Workshop on Robot and Human Communication, RO-MAN (IEEE International Workshop on Robot and Human Communication, RO-MAN) 1196 - 1201 2023

ISSN 19449445

Collaborative delivery robots in hospitals are required to move safely and efficiently in a short time, without colliding with people. Hence, they must consider the risk of people rushing out from blind spots or rooms, including field context such as the role and usage of the location. However, these factors are difficult to extract solely from geometric information. Therefore, we propose a method for generating a rush-out risk map considering the field context from the hospital staf's operation data of an electric wheelchair. We convert the wheelchair's speed operated by staff into rush-out risk, and then place rush-out risk potentials at positions where rush-outs may occur. Subsequently, we optimize the mapping position of rush-out risk and parameters of each potential to minimize the error to obtain a rush-out risk map. We collected actual staff operation data in the hospital and confirmed that we could generate a rush-out risk map with small errors.

Access to Document (DOI)

Physics-Based Motion Planning of a Fruit Harvesting Manipulator for Pushing Obstacles in a Cluttered Environment

Nakamura M., Kawasaki Y., Takahashi M.

Lecture Notes in Networks and Systems (Lecture Notes in Networks and Systems) 577 LNNS 204 - 215 2023

ISSN 23673370

Recently, the agricultural population is decreasing, and the demand for autonomous agricultural robots has increased. For a robot to successfully harvest fruits in diverse environments, various obstacles must be pushed away from its path. During the pushing motion, however, gravity acts on the fruits, causing them to move back and forth and return to their original state after being pushed away. This results in fruit-to-fruit contact, and fruit damage may occur. Therefore, a method to avoid obstacles while reducing the risk of fruit damage is required. However, the direct consideration of fruit damage is difficult because it requires many parameters that are difficult to observe. Herein, we propose a motion planning method by considering the potential energy of fruits as an evaluation index, which could be calculated based on observable parameters; this consideration could also suppress the reciprocating motion of fruits, thus preventing fruit damage. The proposed method was evaluated through simulations, and its effectiveness was verified.

Access to Document (DOI)

Multi-Agent Action Graph Based Task Allocation and Path Planning Considering Changes in Environment

Okubo T., Takahashi M.

IEEE Access (IEEE Access) 11 21160 - 21175 2023

Task allocation and path planning considering changes in the mobility of robots in the environment allows the robots to efficiently execute tasks with smaller travel times. A lunar base construction is one of the situations in which robots can more efficiently accomplish its goal by taking such environment changes into account when performing tasks. For the construction, we assumed that when a robot executes a task of building a road, the environment changes such that aisles that were unusable before the task become usable post execution. If such changes in environment are considered in advance, the robot can efficiently plan to wait until the environment changes and can move before executing the task. However, previous studies have not considered such changes, resulting in inefficient planning. To solve this problem, we developed a multi-agent action graph that consists of multiple layers and expresses the environment changes associated with task execution in terms of changes in these layers. In this graph, task allocation and path planning are formulated as a combinatorial optimization problem and are optimized using mixed-integer programming. Multi-agent action graphs and the proposed formulation enable efficient planning considering changes in the robots' mobility in advance. Through simulations, we confirmed that the proposed method completed the construction of the lunar base approximately 16.4% earlier than the conventional method, while consuming approximately 16.0% less total energy of the robots.

Access to Document (DOI)

Modeling of Body Pressure Distribution on Alternative Pressure Mattress

Kosuge M., Takahashi M.

2023 9th International Conference on Automation, Robotics and Applications, ICARA 2023 (2023 9th International Conference on Automation, Robotics and Applications, ICARA 2023) 289 - 293 2023

Recently, an alternative pressure mattress has been proposed to prevent pressure ulcers. In this study, we aim to design a control system for body pressure distribution using the mattress. This study focuses on modeling the dynamics between the body pressure distribution and the mattress. As modeling based on first principles is challenging, we identify the model from experimental data. However, when high-dimensional data such as body pressure distribution are directly used for identification, the model accuracy can deteriorate. Therefore, we need to determine the suitable model structure or dataset. Herein, we examined how the prediction accuracy of body pressure distribution varies with the dataset and model structure. We also proposed a method for representing the model as the dynamics of the parameters of the Gaussian Mixture Model to suppress the estimated parameters. To the best of our knowledge, this study is the first to propose the body pressure model as dynamics of the parameters of the Gaussian Mixture Model.

Access to Document (DOI)

Learning User-Preferred Robot Navigation Based on Social Force Model from Human Feedback in Virtual Reality Environments

Nakaoka S., Kawasaki Y., Takahashi M.

IEEE International Workshop on Robot and Human Communication, RO-MAN (IEEE International Workshop on Robot and Human Communication, RO-MAN) 1182 - 1187 2023

ISSN 19449445

Autonomous service robots are increasingly necessary to move without impeding the movement of pedestrians. Previous studies have determined optimal input for robots by minimizing a multi-objective function that includes the cost of reaching the destination and avoiding surrounding pedestrians. However, it is challenging to adjust the weights of each term in the cost function since they depend on the users and environment. In this study, we used the Social Force Model (SFM) as the base cost function and proposed a method to estimate SFM weights preferred by general user based on population density from human feedback. To achieve this, first we use Bayesian optimization and derive each user's evaluation map of SFM in a virtual reality environment that provides a realistic and immersive experience for subjects to provide feedback on the robot's movement. Second, we aggregated each user's evaluation map to estimate a general user's evaluation map. Finally, we have derived a functional relationship between the preferred SFM weights of general users and population density by Gaussian process regression. This relationship empowers the robot to navigate in a manner preferred by the general public, contingent on population density, even in the absence of human feedback obtained through virtual reality experimentation.

Access to Document (DOI)

Image-based robot navigation with task achievability

Ishihara Y., Takahashi M.

Frontiers in Robotics and AI (Frontiers in Robotics and AI) 10 2023

Image-based robot action planning is becoming an active area of research owing to recent advances in deep learning. To evaluate and execute robot actions, recently proposed approaches require the estimation of the optimal cost-minimizing path, such as the shortest distance or time, between two states. To estimate the cost, parametric models consisting of deep neural networks are widely used. However, such parametric models require large amounts of correctly labeled data to accurately estimate the cost. In real robotic tasks, collecting such data is not always feasible, and the robot itself may require collecting it. In this study, we empirically show that when a model is trained with data autonomously collected by a robot, the estimation of such parametric models could be inaccurate to perform a task. Specifically, the higher the maximum predicted distance, the more inaccurate the estimation, and the robot fails navigating in the environment. To overcome this issue, we propose an alternative metric, “task achievability” (TA), which is defined as the probability that a robot will reach a goal state within a specified number of timesteps. Compared to the training of optimal cost estimator, TA can use both optimal and non-optimal trajectories in the training dataset to train, which leads to a stable estimation. We demonstrate the effectiveness of TA through robot navigation experiments in an environment resembling a real living room. We show that TA-based navigation succeeds in navigating a robot to different target positions, even when conventional cost estimator-based navigation fails.

Access to Document (DOI)

Identification of Early Knee Osteoarthritis Based on Knee Joint Trajectory during Stair Climbing

Ogawa A., Iijima H., Takahashi M.

International Journal of Environmental Research and Public Health (International Journal of Environmental Research and Public Health) 19 ( 22 ) 2022.11

ISSN 16617827

Patients with knee osteoarthritis show low stair climbing ability, but a diagnosis of stair performance time is not enough to identify the early stages of knee osteoarthritis. Therefore, we developed an indicator named range of the knee joint trajectory (RKJT) as a kinematic parameter to express more detailed characteristics than stair performance time. To achieve this, we used our developed “IR-Locomotion”, a markerless measurement system that can track the knee joint trajectory when climbing stairs. This study aimed to test whether the RKJT effectively identifies patients with early knee osteoarthritis even after controlling stair performance time. Forty-seven adults with moderate to severe knee pain (mean age 59.2 years; 68.1% women) underwent the radiographic examination (Kellgren and Lawrence grade) of both knees and a stair climbing test on 11 stairs. The RKJT during the stair climbing test was calculated by “IR-Locomotion”. A generalized linear mixed model was used to evaluate the discriminative capability of RKJT on early knee osteoarthritis (i.e., Kellgren and Lawrence grade of 1). As expected, patients with early knee osteoarthritis showed larger RKJT than non-radiographic controls (95% confidence interval: 1.007, 1.076). Notably, this finding was consistent even after adjusting stair performance time.

Access to Document (DOI)

Decision-Making Based on Reinforcement Learning and Model Predictive Control Considering Space Generation for Highway On-Ramp Merging

Kimura H., Takahashi M., Nishiwaki K., Iezawa M.

IFAC-PapersOnLine (IFAC-PapersOnLine) 55 ( 27 ) 241 - 246 2022.09

Reducing traffic accidents pertaining to autonomous vehicles has garnered attention. Merging on a highway is one of the most challenging problems that must be addressed for the realization of autonomous vehicles. It is difficult because an agent must decide where to merge in a complex and ever-changing environment. Merging with congested highway traffic involves significant interaction with vehicles in the main lane. If there is no space for the autonomous vehicle to merge, it needs to work on vehicles in the main lane to create space and subsequently decide to merge or not. Reinforcement learning (RL) is a promising method for solving decision-making problems. However, it is difficult to guarantee the safety of the controller obtained using RL. Therefore, we propose a combined method in which decision-making is performed by RL and vehicle control by model predictive control (MPC) to ensure safety. The performance of the proposed system is tested by simulations. The proposed system made appropriate decisions according to the situation, and by controlling the vehicle in consideration of collision avoidance constraints, it showed a high merge success rate evenin a crowded situation.

Access to Document (DOI)

Time-constrained Optimization of Sit-to-stand Movements in Contact with the Environment

Inoue Y., Takahashi M.

IFAC-PapersOnLine (IFAC-PapersOnLine) 55 ( 27 ) 358 - 363 2022.09

This study aims to evaluate effects of different lower limb joint torque constraints on the motion trajectory of an sit-to-stand movement (STS) grasping an assistive device by simulating the STS with an assistive device using a human model with different lower limb joint torque constraints at various second lengths. This is an initial study for estimating lower limb muscle strength from the motion trajectory of the STS grasping an assistive device. In this study, instead of measuring many human subjects, optimal motion trajectories are calculated and simulated. Equations of the STS using the assistive device are first established using a rigid link model with both hands grasping the assistive device. Then, equations of motion and physical constraints such as joint range of motion and torque limitations are used as constraint conditions, and the sum of squares of joint torque, which calculates the action with the minimum muscle load on the body, is used as an evaluation function to optimize the STS. During optimization, calculations are performed with several types of joint torque conditions and end-time, i.e., the time required to complete the operation. By comparing several types of joint torque condition movements, the differences in the effects of different joint torque constraints on the motion trajectory of the STS grasping the assistive device can be determined.

Access to Document (DOI)

Analysis of Spastic Gait in Patients with Cervical Myelopathy Using the Timed up and Go Test with a Laser Range Sensor

Koyama T., Fujita K., Iijima H., Norose M., Ibara T., Sasaki T., Yoshii T., Nimura A., Takahashi M., Okawa A.

Spine (Spine) 47 ( 12 ) 892 - 898 2022.06

ISSN 03622436

Study Design. Cross-sectional study. Objective. This study aimed to objectively evaluate spastic gait and reveal its novel characteristics via analysis of gait in patients with cervical myelopathy (CM) using the Timed Up and Go (TUG) test with a laser range sensor. Summary of Background Data. Among patients with CM, spastic gait is a common diagnostic symptom; thus, objective assessments of spastic gait would be useful for the diagnosis of CM and recognition of disease status. Although spastic gait has been objectively evaluated in previous studies, the methods employed in those studies are not suitable for clinical settings. Methods. In total, 37 and 24 participants were recruited for a control group and CM group, respectively. CM was diagnosed by spine surgeons. We developed a laser TUG test, in which the position and velocity of both the legs were captured. The parameter values for both groups were statistically compared, and odds ratios were calculated using logistic regression analyses. Results. The total TUG-test time, time to stand up, time to first step, number of steps, and trajectory error for the CM group were significantly higher than those for the control group, whereas the average velocity and average stride length for the CM group were significantly lower than those for the control group. There was a significant independent association between the total TUG-test time and CM. The optimal cutoff point of the total test time for CM risk was approximately 9seconds. Conclusion. Through the use of the laser TUG test, we were able to identify characteristics of spastic gait, which leads to difficulty in standing and taking the first step, wobbling while walking, and an increased risk of falling. We found that the risk of CM was higher if the individual took longer than 9seconds to complete the TUG test.

Access to Document (DOI)

Empirical study of future image prediction for image-based mobile robot navigation

Ishihara Y., Takahashi M.

Robotics and Autonomous Systems (Robotics and Autonomous Systems) 150 2022.04

ISSN 09218890

Recent image-based robotic systems use predicted future state images to control robots. Therefore, the prediction accuracy of the future state image affects the performance of the robot. To predict images, most previous studies assume that the camera captures the entire scene and that the environment is static. However, in real robot applications, these assumptions do not always hold. For example, if a camera is attached to a mobile robot, its view changes from time to time. In this study, we analyzed the relationship between the performance of the image prediction model and the robot's behavior, controlled by an image-based navigation algorithm. Through mobile robot navigation experiments using front-faced and omni-directional cameras, we discussed the capabilities of the image prediction models and demonstrated their performance when applied to the image-based navigation algorithm. Moreover, to adapt to the dynamic changes in the environment, we studied the effectiveness of directing the camera to the ceiling. We showed that robust navigation can be achieved without using images from cameras directed toward the front or the floor, because these views can be disturbed by moving objects in a dynamic environment.

Access to Document (DOI)

Human Leg Tracking by Fusion of Laser Range and Insole Force Sensing With Gaussian Mixture Model-Based Occlusion Compensation

Eguchi R., Takahashi M.

IEEE Sensors Journal (IEEE Sensors Journal) 22 ( 4 ) 3704 - 3714 2022.02

ISSN 1530437X

Simultaneous measurement of spatial, temporal, and kinetic gait parameters has become important for monitoring patients with neurological disorders in a hospital. As spatial sensing technologies, laser range sensors (LRSs), which obtain highly accurate two-dimensional distance data over a wide range, have been employed to detect/track both legs during walking. However, when walking on curved trajectories, a continuous occlusion occurs over several sampling steps and produces false tracking. To address this issue in combination with temporal and kinetic sensing, this paper presents a fusion system of LRSs and force-sensing insoles and an occlusion compensation based on probabilistic motion models. First, the system pre-tracks a target person during walking along a straight line and curved paths under different curvatures/directions in a situation without occlusion. Gait cycles of each walking type are divided by foot-grounding obtained from the insoles. Relationships between leg trajectories and traveling directions during the gait cycle are then learned using user-specific Gaussian mixture models (GMMs). When an occlusion occurs in post-tracking, a maximum likelihood (ML) GMM is identified using a joint probability of both legs' trajectories, in accordance with biomechanics that both legs move in a coordinated manner. The ML GMM then compensates for the traveling direction and position of the occluded leg, and the system interpolates/re-tracks its positions to correct the states during occlusion. Experimental results demonstrated that the proposed method (fusion with the insoles, occlusion compensation, and interpolation/re-tracking) significantly enhanced tracking performance during occlusion and estimated leg positions with valid accuracy (errors of under 60 mm).

Access to Document (DOI)

Dynamic Object Removal from Unpaired Images for Agricultural Autonomous Robots

Akada H., Takahashi M.

Lecture Notes in Networks and Systems (Lecture Notes in Networks and Systems) 412 LNNS 641 - 653 2022

ISSN 23673370

Recently, the demand for agricultural autonomous robots has been increasing. Using the technology of vision-based robotic environmental recognition, they can generally follow farmers to support their work activities, such as conveyance of the harvest. However, a major issue arises in that dynamic objects (including humans) often enter images that the robots rely on for environmental recognition tasks. These dynamic objects degrade the performance of image recognition considerably, resulting in collisions with crops or ridges when the robots are following the worker. To address the occlusion issue, generative adversarial network (GAN) solutions can be adopted as they feature a generative capability to reconstruct the area behind dynamic objects. However, precedented GAN methods basically presuppose paired image datasets to train their networks, which are difficult to prepare. Therefore, a method based on unpaired image datasets is desirable in real-world environments, such as a farm. For this purpose, we propose a new approach by integrating the state-of-the-art neural network architecture, CycleGAN, and Mask R CNN. Our system is trained with a human-tracking dataset collected by an agricultural autonomous robot in a farm. We evaluate the performance of our system both qualitatively and quantitatively for the task of human removal in images.

Access to Document (DOI)

Future Image Prediction for Mobile Robot Navigation: Front-Facing Camera Versus Omni-Directional Camera

Ishihara Y., Takahashi M.

Lecture Notes in Networks and Systems (Lecture Notes in Networks and Systems) 412 LNNS 654 - 669 2022

ISSN 23673370

When we perform a task, we select the action by imagining its future consequences. Hence, the ability to predict future states would also be an essential feature for robotic agents because it would allow them to plan effective actions to accomplish given tasks. In this research, we explore an action-conditioned future image prediction model considering its application to navigation tasks for a mobile robot. We investigate the image prediction performance of deep neural network architectures and training strategies with two different camera systems. One camera system is a conventional front-facing camera that has a narrow field of view with high definition, and the other is an omni-directional camera that has a wide field of view with low definition. We compare the performances of prediction models for these two camera systems, and propose using an image prediction model with the omni-directional camera for the navigation tasks of a robot. We evaluate the prediction performance of each camera system through experiments conducted in a complex living room-like environment. We demonstrate that models with an omni-directional camera system outperform models with a conventional front-facing camera. In particular, the model comprising a combination of action-conditioned long short-term memory successfully predicts future images for states of more than 100 steps ahead in both simulation and real-world scenarios. Further, by integrating the proposed system into image-prediction-based navigation algorithm, we demonstrate that navigation based on a model with an omni-directional camera can successfully navigate the robot in cases where one with a conventional front-facing camera fails.

Access to Document (DOI)

　 Previous page - Next page 　