Showing papers by "Keigo Watanabe published in 2009"

Biped locomotion using CPG with sensory interaction

Abstract: This article presents the design and analysis of a controller based on a biologically inspired central pattern generator (CPG) network of mutually coupled Matsuoka nonlinear oscillators to generate adaptive rhythmic human like movement for biped robots The paper focuses on the way in which the sensory signals feedback contribute to generate dynamic, stable and sustained rhythmic movements with robust gaits for biped robots In addition, the paper shows how the driving input and external perturbation affect the speed of locomotion and change the period of its own active phase The new design was studied through interaction between simulated interconnection coupling dynamics with 6 links and a musculoskeletal model with 6 degrees of freedom (DOFs) of a biped robot The robot used the weighted outputs of mutually inhibited oscillators as torques to actuate its joints The implemented model helps to realize the interaction between the controller, the mechanism of the robot, and the environment In addition, it helps to study the necessary conditions for efficient generation of stable rhythmic walking at different speed, on different type of terrains and robustness in response to disturbances Evaluations of the developed CPG based adaptive bipedal locomotion are carried out through simulations with successful testing results

Intelligent machining system for the artistic design of wooden paint rollers

Abstract: In this paper, a 3D design and machining system based on a 3-axis NC machine tool with a rotary unit is introduced to efficiently produce the artistic design of wooden paint rollers. The paint rollers are used to execute a relief wall just after painting. A simple post-processor is first proposed for the NC machine tool to transform a base tool path called cutter location data (CL data) to NC data, mapping the y-directional pick feed to the rotational angle of the rotary unit. Also, the post-processor has a novel function that elaborately adjusts feed rate values according to the curvature of each design so as not to chip the carved surface. The suitable feed rate values are generated by using a simple fuzzy reasoning method while checking edges and curvatures in a relief design. The post-processor allows the 3-axis NC machine tool with a rotary unit to easily carve an artistic relief design on a cylindrical wooden workpiece without an undesirable edge chipping. Experimental results show that wooden paint rollers with an artistic relief design can be successfully machined without any chipping.

Discontinuous Control and Backstepping Method for the Underactuated Control of VTOL Aerial Robots with Four Rotors

Abstract: A control strategy is proposed here for four-rotor vertical take-off and landing (VTOL) aerial robot called X4-flyer. Since the X4-flyer has underactuated and nonholonomic features, a kinematics control law is first derived using Astolfi’s discontinuous control. A backstepping method that is one of adaptive control methods based on Lyapunov methods, then provides the kinematic based inputs, to construct the torque control of X4-flyer. Finally, computer simulations are given to demonstrate the effectiveness of our approach.

Adaptation of robot behaviors toward user perception on fuzzy linguistic information by fuzzy voice feedback

Abstract: This paper proposes a method to adapt robot behaviors toward user's perception by human teaching. Human-friendly robotic system should be able to understand the fuzzy linguistic information based on the user's guidance and the environmental conditions. The contextual meaning of fuzzy linguistic information depends on the conditions of the environment. Therefore, user's perception is acquired to evaluate the fuzzy linguistic information in user commands based on fuzzy voice feedback. The primitive behaviors are evaluated by behavior evaluation network (BEN). Feedback evaluation system (FES) is introduced to evaluate the user feedback to correct the robot's perception by adapting the BEN. This yields the adaptation of the system for understanding fuzzy linguistic information toward the corresponding environment. A situation of cooperative rearrangement of user's working space is simulated to illustrate the system. This is demonstrated by using a PA-10 robot manipulator.

Understanding user commands by evaluating fuzzy linguistic information based on visual attention

Abstract: This article proposes a method for understanding user commands based on visual attention. Normally, fuzzy linguistic terms such as “very little” are commonly included in voice commands. Therefore, a robot’s capacity to understand such information is vital for effective human-robot interaction. However, the quantitative meaning of such information strongly depends on the spatial arrangement of the surrounding environment. Therefore, a visual attention system (VAS) is introduced to evaluate fuzzy linguistic information based on the environmental conditions. It is assumed that the corresponding distance value for a particular fuzzy linguistic command depends on the spatial arrangement of the surrounding objects. Therefore, a fuzzy-logic-based voice command evaluation system (VCES) is proposed to assess the uncertain information in user commands based on the average distance to the surrounding objects. A situation of object manipulation to rearrange the user’s working space is simulated to illustrate the system. This is demonstrated with a PA-10 robot manipulator.

A Two-Step Approach to Extracting Attributes for People on the Web

Abstract: Personal names are among one of the most frequently searched items in web search engines. Extracting information in the form of attributes and values for a particular person enables us to uniquely identify that person on the web. For example, although namesakes share the same name they usually have different date of births or affiliations. Given a set of documents retrieved for a particular person, we propose two stage approach to extract values for a set of attributes for that person. In the first stage we mark all potential attribute strings in a given text. The second stage then attempts to select the attribute values relevant to a person name. We use a named entity recognition tool to mark all occurrences of named entities in a given document. We then use a rule-based tagger to identify the variants of the given person name. Next, we employ a combination of rules and pre-compiled attribute value candidate lists to extract values for a given set of attributes. The candidate value lists are manually created using resources available on the web such as Wikipedia. The proposed method is evaluated on the test data collection created for the attribute extraction subtask at the second Web People Search Task (WePS). According to the results in the official evaluation, the proposed method is ranked 5-th among the 15 participating systems.

Intelligent desktop NC machine tool with compliant motion capability

Abstract: In this paper, a new desktop NC machine tool with compliance control capability is presented for finishing metallic molds with small curved surface. The NC machine tool consists of three single-axis robots. Tools attached to the tip of the z-axis are ball-end abrasive tools. The control system of the NC machine tool is composed of a force feedback loop, position feedback loop and position feed-forward loop. The force feedback loop controls the polishing force consisting of tool contact force and kinetic friction force. The position feedback loop controls the position in pick feed direction. Further, the position feed-forward loop leads the tool tip along cutter location data. In order to first confirm the application limit of a conventional industrial robot to a finishing task, we evaluate the backlash that causes the position inaccuracy at the tip of an abrasive tool, through a simple position/force measurement. Through a similar measurement and a surface following control experiment along a lens mold, the basic position/force controllability with high resolutions is demonstrated.

Voice instructions for controlling a robotic forceps with image and auxiliary information

Abstract: An image related to three-dimensional movement is used for constructing a control system for a robotic forceps by voice instructions, together with use of auxiliary information such as, the information on the tip-position in three-dimension and on the distance of the previous movement. A degree-related adverb included in the voice instruction is quantified by using fuzzy reasoning, in which such auxiliary information is also applied to update the membership functions so as to realize a user-friendly interface. The effectiveness of the present system is demonstrated by actual experiments.

Behavior generation in robots by emotional motivation

Abstract: Emotions are necessary in robotic system to get some extra benefits especially in intelligence and autonomy with the motivational mechanism for behavior generation. A robot needs to be intelligent enough to adjust with dynamic and flexible environment. Benefits can be amplified if emotional intelligence can be incorporated with the available control methods. We describe an emotion model to be used for a robotic system with an application in a simulated environment. The simulation shows satisfactory results in emotion generation and actions selection.

Evaluating fuzzy voice commands by internal rehearsal for controlling a robot manipulator

Abstract: This paper proposes a method for evaluating fuzzy voice commands (FVCs) by internal rehearsal. The robot's experience on previous movements is very useful for understanding fuzzy linguistic information. Therefore, an internal rehearsal system is introduced to acquire the robot's experience on previous movements. Such acquisition enhances the robot's capability to interpret fuzzy linguistic information in FVCs according to the current environmental context. The proposed system is used to evaluate the fuzzy linguistic information related to the primitive movements of a robot manipulator. Those primitive movements are realized by behavior evaluation network (BEN) with the guidance of the internal rehearsal system. Mainly, the system is utilized to understand fuzzy voice motion commands (FVMCs) and fuzzy voice joint commands (FVJCs) for controlling the end-effector movements and the single joint movements of the robot manipulator. The proposed idea is demonstrated with a PA-10 robot manipulator.

Evaluating fuzzy voice commands by user guided visual attention for controlling a robot manipulator

Abstract: This paper proposes a method for evaluating fuzzy voice commands (FVCs) by user guided visual attention. FVCs inherit fuzzy linguistic information due to the inclusion of fuzzy linguistic terms such as “very little.” The contextual meaning of such information depends on the environmental arrangement. Therefore, visual attention system (VAS) is utilized to evaluate the fuzzy linguistic information based on the environmental conditions. It is implemented by assuming that the corresponding distance value for a particular FVC depends on the spatial arrangement of the surrounding objects in the environment. A fuzzy logic based voice command evaluation system (VCES) is proposed to assess the fuzzy linguistic information in user commands. In addition, the user has been given the freedom to adjust command evaluation process by changing the attention level. It is achieved by using attentive voice commands such as “move carefully.” A situation of object manipulation by navigating the tip of a robot manipulator through the user's working space is demonstrated to illustrate the system where a PA-10 robot manipulator is used.

Multi-robot team work with benevolent characters: The roles of emotions

Abstract: Multi-robot team work is necessary for complex tasks which cannot be performed by a single robot. To get the required performance and reliability, it is necessary to develop a proper cooperative task. The robots need to be intelligent enough to adjust with dynamic workload and environment. Benefits can be amplified from a team if benevolence combines with cooperation. The benevolence behaviors among the team members are extra benefits to the society. There is a flexible relation among intelligence, benevolence and emotions. We describe an emotion model to be used for each of the members of a multi-robot team. In respect of some drawbacks with the existing approaches, we present an emotion based multi-robot cooperation with some benevolent characters.

Kinematics-based control of underactuated vehicles with four-inputs and six-states by applying invariant manifolds

Abstract: In this paper, an invariant manifold approach is applied for controlling chained form systems with four inputs and six states. This type of systems includes VTOL (vertical take off and landing) aerial robots such as X4-flyer, AUVs (autonomous underwater vehicles) such as a longitudinal or lateral X4 -AUV, etc. The present underactuated control method is composed of a two step method: the first step is an attractive control to achieve an invariant manifold vector and the second step is a state feedback control to assure the asymptotic behavior of states constrained on each invariant manifold. Some simulations with changes in design parameters are given to demonstrate the effectiveness of the proposed method.

Control of three-link underactuated manipulators by a logic-based switching method

Abstract: Control of underactuated manipulators is an attractive research theme in robotics. One of the reasons is that they become a complex system, so that they cannot be controlled by normal any control methods. These systems also include the properties of nonholonomic systems. One of control methods used for underactuated manipulators is known as a switching control, which selects a stable controller out of ones prepared in advance by using a switching rule. In this paper, we propose a switching control method for three-DOF underactuated manipulators. The dynamical model of an underactuated manipulator consisting of two prismatic joints and one revolute joint is first transformed into an extended nonholonomic double integrator form. Such a system can be controlled by applying the conventional logic-based control method developed in the case of velocity constraint, after slightly extending it to the present case. The effectiveness of the proposed method is illustrated through simulations with a three-DOF underactuated manipulator.

Discontinuous Control and Backstepping Method for the Underactuated Control of VTOL Aerial Robots with Four Rotors.

Abstract: A control strategy is proposed here for four-rotor vertical take-off and landing (VTOL) aerial robot called X4-flyer. Since the X4-flyer has underactuated and nonholonomic features, a kinematics control law is first derived using Astolfi’s discontinuous control. A backstepping method that is one of adaptive control methods based on Lyapunov methods, then provides the kinematic based inputs, to construct the torque control of X4-flyer. Finally, computer simulations are given to demonstrate the effectiveness of our approach.

Posture control of a robot manipulator by evaluating fuzzy linguistic information based on user feedback

Abstract: This paper proposes a method for controlling posture of a robot manipulator by fuzzy voice commands. Human-friendly robotic system should be able to understand the fuzzy linguistic information based on the user's guidance and the environmental conditions. The contextual meaning of fuzzy linguistic information depends on the conditions of the environment. Therefore, the user's feedback is evaluated to understand the fuzzy linguistic information related to the posture movements. The primitive posture movements are evaluated by the behavior evaluation network (BEN). Feedback evaluation system (FES) is introduced to evaluate the user's feedback to correct the robot perception by adapting the BEN. The capability of evaluating fuzzy linguistic information based on the current context is enhanced. A selected set of posture movements are used to illustrate the system by using a PA-10 robot manipulator.

Impedance model force control using neural networks for a desktop NC machine tool

Abstract: In manufacturing industries of metallics molds, various NC machine tools are used. We have already proposed a desktop NC machine tool with compliance control capability to automatically cope with the finishing process of LED lens molds. The NC machine tool has an ability to control the polishing force acting between an abrasive tool and workpiece. The force control method is called impedance model force control. The most important gain is the desired damping of the impedance model. Ideally, the desired damping is calculated from the critical damping condition in consideration of the effective stiffness in force control system. However, one of the serious problems is that the effective stiffness of the NC machine tool has undesirable nonlinearity. The nonlinearity gives bad influences to the force control stability. In this paper, a fine tuning method of the desired damping is considered by using neural networks. The neural networks acquire the nonlinearity of effective stiffness. It has been observed that the desired damping generated from the learned neural networks allows the NC machine tool to achieve a stable finishing result.

Switching Control of Underactuated Manipulators using Fuzzy Energy Regions

Abstract: 本稿では非ホロノミックの特性を有する劣駆動マニピュレータについて議論する．劣駆動マニピュレータは関節数よりも少ないアクチュエータを装備しており，関節の一部には非駆動関節を持つ．この非駆動関節の影響により通常のマニピュレータよりも複雑な制御を必要とし，問題解決のために現在多くの研究が行われている．また劣駆動マニピュレータを適用した場合に考えられる利点としては，アクチュエータ数を減少させることによる軽量化や省エネルギー化，コスト削減等が挙げられる．本研究では劣駆動マニピュレータを制御するために各リンクに応じたエネルギー領域を構成し，論理型切換え法でのその境界曲線にファジィルール表現を導入することにより，あいまい性を含ませ制御器の切換えに柔軟性を持たせる，ファジィエネルギー領域切換え法を考案している．制御対象である３リンク２アクチュエータの劣駆動マニピュレータについて，エネルギー領域を２次元平面で構成する場合，複数の組み合わせが存在する．ここでは２次元面でのエネルギーの射影形態と制御性能について考察し，各組み合わせに対するシミュレーション結果を示す．

Orthogonal-type robot with a CAD/CAM-based position/force controller

Abstract: In this paper, a new desktop orthogonal-type robot with a CAD/CAM-based position/force controller is presented for finishing small metallic molds with curved surface. The robot consists of a three-axis robot whose single one has a high position accuracy of 1 µm, which means that it can perform higher resolutions of position and force, compared to general industrial robots. A thin wood stick tool with a ball-end tip is attached to the z-axis through a force sensor. The control system of the robot is comprised of a force feedback loop, position feedback loop and position feedforward loop. The force feedback loop controls the polishing force consisting of tool contact force and kinetic friction forces. The position feedback loop controls the position in pick feed direction. Further, the position feedforward loop leads the tool tip along a spiral path. We first evaluate the backlash that causes inaccuracy in position at the tip of the abrasive tool, by simply measuring the position and force. Next, a surface following control experiment along a lens mold is conducted, in which the mold has axis-symmetric concave areas. Finally, a LED lens mold is further finished by using the proposed system in order to demonstrate the performance and promise.