CAMBADAsoccerteam:fromrobotarchitecturetomultiagentcoordination 43 distributed architectures extend from improved composability, allowing a system to be built by putting together different subsystems, to higher scalability, allowing to add functionality to the system by adding more nodes, more flexibility, allowing to reconfigure the system easily, better maintainability, due to the architecture modularity and easiness of node replacement, and higher reduction of mutual interference, thus offering a strong potential to support reac- tive behaviors more efficiently. Moreover, distributed architectures may also provide benefits in terms of dependability by creating error-containment regions at the nodes and opening the way for inexpensive spatial replication and fault tolerance. The vision system of the CAMBADA robots is based on an hybrid system, formed by an om- nidirectional and a perspective sub-system, that together can analyze the environment around the robots, both at close and long distances. We presented in this chapter several algorithms for the calibration of the most important parameters of the vision system and we propose efficient color-based algorithms for object detection. Moreover, we proposed a promising so- lution for the detection of arbitrary FIFA balls, as demonstrated by the first place obtained in the mandatory technical challenge in the Robocup 2009, where the robots had to play with an arbitrary standard FIFA ball. Sensor and information fusion is important to maintain a more reliable description of the state of the world. The techniques proposed for information and sensor fusion proved to be effec- tive in accomplishing their objectives. The Kalman filter allows to filter the noise on the ball position and provides an important prediction feature which allows fast detection of devia- tions of the ball path. The linear regression used to estimate the velocity is also effective, and combined with the deviation detection based on the Kalman filter prediction error, provides a faster way to recalculate the velocity in the new trajectory. The increasing reliability of the ball position and velocity lead to a better ball trajectory evaluation. This allowed the devel- opment of a more effective goalie action, as well as other behaviors, such as ball interception behaviors and pass reception. The results regarding obstacle identification, provide tools for an improvement of the overall team coordination and strategic play. The robots coordination is based on a replicated database, the Real-Time Data Base (RTDB) that includes local state variables together with images of remote ones. These images are up- dated transparently to the application software by means of an adequate real-time manage- ment system. Moreover, the RTDB is accessible to the application using a set of non-blocking primitives, thus yielding a fast data access. Regarding the communication between robots, is was developed a wireless communication protocol that reduces the probability of collisions among the team members. The protocol called adaptive TDMA, adapts to the current channel conditions, particularly accommodating periodic interference patterns. It was also developed an extended version of the protocol with on-line self-configuration capabilities that allow reconfiguring the slots structure of the TDMA round to the actual number of active team members, further reducing the collision probability. In the CAMBADA MSL team, each robot is an independent agent and coordinates its actions with its teammates through communication and information exchange. The resulting behav- ior of the individual robot should be integrated into the global team strategy, thus resulting in cooperative actions by all the robots. This is done by the use of roles and behaviors that define each robot attitude in the field and resulting individual actions. This resulted in a coordinated behavior of the team that largely contributed to its recent successes. The base station application has a important role during the development of a robotic soccer team and also during a game. This chapter presented the approach that was used by the CAMBADA team in the design of this important application. The CAMBADA MSL team attained the first place in the MSL at RoboCup 2008 and attained the third place in the last edition of the MSL at RoboCup 2009. CAMBADA also won the last three editions of the Portuguese Robotics Open 2007-2009. These results confirm the effective- ness of the proposed architecture. 10. References Almeida, L., P. Pedreiras, and J. A. Fonseca (2002). The FTT-CAN protocol: Why and how. IEEE Transactions on Industrial Electronics 49(6), 1189–1201. Almeida, L., F. Santos, T. Facchinetti, P. Pedreira, V. Silva, and L. S. Lopes (2004). Coordinating distributed autonomous agents with a real-time database: The CAMBADA project. In Proc. of the 19th International Symposium on Computer and Information Sciences, ISCIS 2004, Volume 3280 of Lecture Notes in Computer Science, pp. 878–886. Springer. Azevedo, J. L., B. Cunha, and L. Almeida (2007). Hierarchical distributed architectures for au- tonomous mobile robots: a case study. In Proc. of the 12th IEEE Conference on Emerging Technologies and Factory Automation, ETFA2007, Greece, pp. 973–980. Bishop, G. and G. Welch (2001). An introduction to the kalman filter. In Proc of SIGGRAPH, Course 8, Number NC 27599-3175, Chapel Hill, NC, USA. Blaffert, T., S. Dippel, M. Stahl, and R. Wiemker (2000). The laplace integral for a watershed segmentation. In Proc. of the International Conference on Image Processing, 2000, Vol- ume 3, pp. 444–447. Caleiro, P. M. R., A. J. R. Neves, and A. J. Pinho (2007, June). Color-spaces and color segmen- tation for real-time object recognition in robotic applications. Revista do DETUA 4(8), 940–945. Canny, J. F. (1986, November). A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 8(6). Cunha, B., J. L. Azevedo, N. Lau, and L. Almeida (2007). Obtaining the inverse distance map from a non-SVP hyperbolic catadioptric robotic vision system. In Proc. of the RoboCup 2007, Atlanta, USA. Ferrein, A., L. Hermanns, and G. Lakemeyer (2006). Comparing sensor fusion techniques for ball position estimation. In RoboCup 2005: Robot Soccer World Cup IX, Volume 4020 of LNCS, pp. 154–165. Springer. Grimson, W. E. L. and D. P. Huttenlocher (1990). On the sensitivity of the hough transform for object recognition. IEEE Trans. on Pattern Analysis and Machine Intelligence 12, 1255– 1274. Lau, N., L. S. Lopes, and G. Corrente (2008, April). CAMBADA: information sharing and team coordination. In Proc. of the 8th Conference on Autonomous Robot Systems and Competitions, Portuguese Robotics Open - ROBOTICA’2008, Aveiro, Portugal, pp. 27– 32. Lauer, M., S. Lange, and M. Riedmiller (2005). Modeling moving objects in a dynamically changing robot application. In KI 2005: Advances in Artificial Intelligence, Volume 3698 of LNCS, pp. 291–303. Springer. Lauer, M., S. Lange, and M. Riedmiller (2006). Calculating the perfect match: an efficient and accurate approach for robot self-localization. In RoboCup 2005: Robot Soccer World Cup IX, Lecture Notes in Computer Science, pp. 142–153. Springer. Marcelino, P., P. Nunes, P. Lima, and M. I. Ribeiro (2003). Improving object localization through sensor fusion applied to soccer robots. In Proc. Scientific Meeting of the Por- tuguese Robotics Open - ROBOTICA 2003. RobotSoccer44 Martins, D. A., A. J. R. Neves, and A. J. Pinho (2008, october). Real-time generic ball recogni- tion in RoboCup domain. In Proc. of the 11th edition of the Ibero-American Conference on Artificial Intelligence, IBERAMIA 2008, Lisbon, Portugal. Motulsky, H. and A. Christopoulos (2003). Fitting models to biological data using linear and nonlinear regression. GraphPad Software Inc. MSL Technical Committee 1997-2009 (2008). Middle size robot league rules and regulations for 2009. Neves, A. J. R., A. J. P. B. Cunha, and I. Pinheiro (2009, June). Autonomous configuration of parameters in robotic digital cameras. In Proc. of the 4th Iberian Conference on Pattern Recognition and Image Analysis, ibPRIA 2009, Póvoa de Varzim, Portugal. Neves, A. J. R., G. Corrente, and A. J. Pinho (2007). An omnidirectional vision system for soccer robots. In Progress in Artificial Intelligence, Volume 4874 of Lecture Notes in Artificial Inteligence, pp. 499–507. Springer. Neves, A. J. R., D. A. Martins, and A. J. Pinho (2008, April). A hybrid vision system for soccer robots using radial search lines. In Proc. of the 8th Conference on Autonomous Robot Systems and Competitions, Portuguese Robotics Open - ROBOTICA’2008, Aveiro, Portugal, pp. 51–55. Reis, L., N. Lau, and E. Oliveira (2001). Situation based strategic positioning for coordinat- ing a team of homogeneous agents. In Balancing Reactivity and Social Deliberation in Multi-Agent Systems, Volume 2103 of Lecture Notes in Computer Science, pp. 175–197. Springer. Santos, F., L. Almeida, L. S. Lopes, J. L. Azevedo, and M. B. Cunha (2009). Communicating among robots in the robocup middle-size league. In RoboCup 2009: Robot Soccer World Cup XIII, Lecture Notes in Artificial Intelligence. Springer (In Press). Santos, F., G. Corrente, L. Almeida, N. Lau, and L. S. Lopes (2007, December). Self- configuration of an adaptive TDMA wireless communication protocol for teams of mobile robots. In Proc. of the 13th Portuguese Conference on Artificial Intelligence, EPIA 2007, Guimarães, Portugal. Ser, P K. and W C. Siu (1993). Invariant hough transform with matching technique for the recognition of non-analytic objects. In IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 1993., Volume 5, pp. 9–12. Silva, J., N. Lau, J. Rodrigues, and J. L. Azevedo (2008, october). Ball sensor fusion and ball interception behaviours for a robotic soccer team. In Proc. of the 11th edition of the Ibero-American Conference on Artificial Intelligence, IBERAMIA 2008, Lisbon, Portugal. Silva, J., N. Lau, J. Rodrigues, J. L. Azevedo, and A. J. R. Neves (2009). Sensor and information fusion applied to a robotic soccer team. In RoboCup 2009: Robot Soccer World Cup XIII, Lecture Notes in Artificial Intelligence. Springer (In Press). Silva, V., R. Marau, L. Almeida, J. Ferreira, M. Calha, P. Pedreiras, and J. Fonseca (2005). Im- plementing a distributed sensing and actuation system: The CAMBADA robots case study. In Proc. of the 10th IEEE Conference on Emerging Technologies and Factory Automa- tion, ETFA2005, Italy, pp. 781–788. Stone, P. and M. Veloso (1999). Task decomposition, dynamic role assignment, and low- bandwidth communication for real-time strategic teamwork. Volume 110, pp. 241– 273. XU, Y., C. JIANG, and Y. TAN (2006). SEU-3D 2006 Soccer simulation team description. In CD Proc. of RoboCup Symposium 2006. Zhang, Y J. and Z Q. Liu (2000). Curve detection using a new clustering approach in the hough space. In IEEE International Conference on Systems, Man, and Cybernetics, 2000, Volume 4, pp. 2746–2751. Zin, T. T., H. Takahashi, and H. Hama (2007). Robust person detection using far infrared camera for image fusion. In Second International Conference on Innovative Computing, Information and Control, ICICIC 2007, pp. 310–310. Zou, J., H. Li, B. Liu, and R. Zhang (2006). Color edge detection based on morphology. In First International Conference on Communications and Electronics, ICCE 2006, pp. 291–293. Zou, Y. and W. Dunsmuir (1997). Edge detection using generalized root signals of 2-d median filtering. In Proc. of the International Conference on Image Processing, 1997, Volume 1, pp. 417–419. CAMBADAsoccerteam:fromrobotarchitecturetomultiagentcoordination 45 Martins, D. A., A. J. R. Neves, and A. J. Pinho (2008, october). Real-time generic ball recogni- tion in RoboCup domain. In Proc. of the 11th edition of the Ibero-American Conference on Artificial Intelligence, IBERAMIA 2008, Lisbon, Portugal. Motulsky, H. and A. Christopoulos (2003). Fitting models to biological data using linear and nonlinear regression. GraphPad Software Inc. MSL Technical Committee 1997-2009 (2008). Middle size robot league rules and regulations for 2009. Neves, A. J. R., A. J. P. B. Cunha, and I. Pinheiro (2009, June). Autonomous configuration of parameters in robotic digital cameras. In Proc. of the 4th Iberian Conference on Pattern Recognition and Image Analysis, ibPRIA 2009, Póvoa de Varzim, Portugal. Neves, A. J. R., G. Corrente, and A. J. Pinho (2007). An omnidirectional vision system for soccer robots. In Progress in Artificial Intelligence, Volume 4874 of Lecture Notes in Artificial Inteligence, pp. 499–507. Springer. Neves, A. J. R., D. A. Martins, and A. J. Pinho (2008, April). A hybrid vision system for soccer robots using radial search lines. In Proc. of the 8th Conference on Autonomous Robot Systems and Competitions, Portuguese Robotics Open - ROBOTICA’2008, Aveiro, Portugal, pp. 51–55. Reis, L., N. Lau, and E. Oliveira (2001). Situation based strategic positioning for coordinat- ing a team of homogeneous agents. In Balancing Reactivity and Social Deliberation in Multi-Agent Systems, Volume 2103 of Lecture Notes in Computer Science, pp. 175–197. Springer. Santos, F., L. Almeida, L. S. Lopes, J. L. Azevedo, and M. B. Cunha (2009). Communicating among robots in the robocup middle-size league. In RoboCup 2009: Robot Soccer World Cup XIII, Lecture Notes in Artificial Intelligence. Springer (In Press). Santos, F., G. Corrente, L. Almeida, N. Lau, and L. S. Lopes (2007, December). Self- configuration of an adaptive TDMA wireless communication protocol for teams of mobile robots. In Proc. of the 13th Portuguese Conference on Artificial Intelligence, EPIA 2007, Guimarães, Portugal. Ser, P K. and W C. Siu (1993). Invariant hough transform with matching technique for the recognition of non-analytic objects. In IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 1993., Volume 5, pp. 9–12. Silva, J., N. Lau, J. Rodrigues, and J. L. Azevedo (2008, october). Ball sensor fusion and ball interception behaviours for a robotic soccer team. In Proc. of the 11th edition of the Ibero-American Conference on Artificial Intelligence, IBERAMIA 2008, Lisbon, Portugal. Silva, J., N. Lau, J. Rodrigues, J. L. Azevedo, and A. J. R. Neves (2009). Sensor and information fusion applied to a robotic soccer team. In RoboCup 2009: Robot Soccer World Cup XIII, Lecture Notes in Artificial Intelligence. Springer (In Press). Silva, V., R. Marau, L. Almeida, J. Ferreira, M. Calha, P. Pedreiras, and J. Fonseca (2005). Im- plementing a distributed sensing and actuation system: The CAMBADA robots case study. In Proc. of the 10th IEEE Conference on Emerging Technologies and Factory Automa- tion, ETFA2005, Italy, pp. 781–788. Stone, P. and M. Veloso (1999). Task decomposition, dynamic role assignment, and low- bandwidth communication for real-time strategic teamwork. Volume 110, pp. 241– 273. XU, Y., C. JIANG, and Y. TAN (2006). SEU-3D 2006 Soccer simulation team description. In CD Proc. of RoboCup Symposium 2006. Zhang, Y J. and Z Q. Liu (2000). Curve detection using a new clustering approach in the hough space. In IEEE International Conference on Systems, Man, and Cybernetics, 2000, Volume 4, pp. 2746–2751. Zin, T. T., H. Takahashi, and H. Hama (2007). Robust person detection using far infrared camera for image fusion. In Second International Conference on Innovative Computing, Information and Control, ICICIC 2007, pp. 310–310. Zou, J., H. Li, B. Liu, and R. Zhang (2006). Color edge detection based on morphology. In First International Conference on Communications and Electronics, ICCE 2006, pp. 291–293. Zou, Y. and W. Dunsmuir (1997). Edge detection using generalized root signals of 2-d median filtering. In Proc. of the International Conference on Image Processing, 1997, Volume 1, pp. 417–419. RobotSoccer46 Small-sizeHumanoidSoccerRobotDesignforFIRAHuroSot 47 Small-sizeHumanoidSoccerRobotDesignforFIRAHuroSot Ching-ChangWong,Chi-TaiCheng,Kai-HsiangHuang,Yu-TingYang,Yueh-YangHuand Hsiang-MinChan X Small-size Humanoid Soccer Robot Design for FIRA HuroSot Ching-Chang Wong, Chi-Tai Cheng, Kai-Hsiang Huang, Yu-Ting Yang, Yueh-Yang Hu and Hsiang-Min Chan Department of Electrical Engineering, Tamkang University Tamsui, Taipei, 25137, Taiwan 1. Introduction Robot soccer games are used to encourage the researches on the robotics and artificial intelligence. FIRA (URL: http://www.fira.net) is an international robot soccer association to advance this research and hold some international competitions and congresses. There are many different leagues, such as SimuroSot, MiroSot, RoboSot, and HuroSot, in FIRA RoboWorld Cup. Each league is established for different research purposes. In the HuroSot league, many technology issues and scientific areas must be integrated to design a humanoid robot. The research technologies of mechanism design, electronic system, biped walking control, autonomous motion, direction judgment, kicking ball need to be applied on a humanoid robot (Chemori & Loria, 2004; Esfahani & Elahinia, 2007; Guan et al., 2006; Hemami et al., 2006; Hu et al., 2008; Haung et al., 2001; Miyazaki & Arimoto, 1980; Sugihara et al., 2002; Wong et al., 2005; Zhou & Jagannathan, 1996). This chapter introduces an autonomous humanoid robot, TWNHR-IV (Taiwan Humanoid Robot-IV), which is able to play sports, such as soccer, basketball, weight lifting, and marathon. The robot is designed to be a vision-based autonomous humanoid robot for HuroSot League of FIRA Cup. TWNHR-IV joined FIRA Cup in 2007 and 2008. In FIRA 2007, TWNHR-IV won the first place in robot dash, penalty kick, obstacle run, and weight lifting; the second place in basketball and marathon. In FIRA 2008, TWNHR-IV won the first place in penalty kick, obstacle run, and weight lifting, the second place in robot dash and the third place in basketball. TWNHR-IV determines the environment via its sensors and executes the suitable motion by its artificial intelligent. In order to let TWNHR-IV have the environment perceptive ability, a vision sensor (a CMOS sensor), six distance sensors (six infrared sensors), a posture sensor (an accelerometer sensor) and a direction sensor (a digital compass) are equipped on the body of TWNHR-IV to obtain the information of the environment. Two processors are used to control the robot. The first one is a DSP for the high-level control purpose. The DSP receives and processes the image data from the CMOS sensor via a serial port. It is in charge of the high level artificial intelligent, such as navigation. The second one is NIOS II (an embedded soft-core processor) for the robot locomotion control. The second processor is used as a low-level controller to control the walking and other actions. TWNHR-IV is designed as a soccer player so that it can walk, 3 RobotSoccer48 turn, and kick the ball autonomously. A control board with a FPGA chip and a 64 Mb flash memory are mainly utilized to control the robot. Many functions are implemented on this FPGA chip so that it can receive motion commands from DSP via a serial port and process the data obtained by infrared sensors, digital compass, and accelerometer. It is able to accomplish the different challenges of HuroSot, including Penalty Kick (PK), basketball, lift- and-carry, obstacle run, robot dash, weight lifting, and marathon autonomously and effectively. The rest of this chapter is organized as follows: In Section 2, the mechanical system design of the robot TWNHR-IV is described. In Section 3, the electronic system including a vision system, a control core, and sensor systems are described. In Section 4, some simulation and experiments results of the proposed controller are described. Finally, some conclusions are made in Section 5. 2. Mechanical System Design Mechanical system design is the first step of design a humanoid robot. Human body mechanism basically consists of bones, joints, muscles, and tendons. It is impossible to replace all of the muscular-skeletal system by current mechanical and electrical components. Therefore, the primary goal of the humanoid robot mechanical system design is to develop a robot that can imitate equivalent human motions. The degrees of freedom (DOFs) configuration of TWNHR-IV is presented in Figure 1. TWNHR-IV has 26 DOFs. In this chapter, the rotational direction of each joint is defined by using the inertial coordinate system fixed on the ground as shown in Figure 1 ( Wong et al., 2008c). Fig. 1. DOFs configuration A photograph and a 3D mechanical structure of the implemented robot are shown in Figure 2 The design concepts of TWNHR-IV are light weight and compact size. The height of TWNHR-IV is 43 cm and the weight is 3.4 kg with batteries. A human-machine interface is designed to manipulate the servo motors. The details of the mechanical structure of the head, arms, waist, trunk, and legs are described as follows. Fig. 2. Photograph and mechanical structure of TWNHR-IV The head of TWNHR-IV has 2 DOFs. Figure 3 shows the 3D mechanism design of the head. The head is designed based on the concept that the head of the robot can accomplish the raw and pitch motion. Table 1 presents the head DOFs relation between human and TWNHR-IV. Human Figure TWNHR-IV Human Figure TWNHR-IV Table 1. The head DOFs relation between human and TWNHR-IV (a) Head 3D design (b) DOFs diagram Fig. 3. Head mechanism The head of TWNHR-IV has 2 degrees of freedom. Figure 4 shows the 3D mechanism design of the waist and trunk. The trunk is designed based on the concept that robot can walk and maintain its balance by using gyro to adjust the trunk motions to compensate for the robot’s walk motion. Table 2 presents the specification of the joints for the waist and trunk. Small-sizeHumanoidSoccerRobotDesignforFIRAHuroSot 49 turn, and kick the ball autonomously. A control board with a FPGA chip and a 64 Mb flash memory are mainly utilized to control the robot. Many functions are implemented on this FPGA chip so that it can receive motion commands from DSP via a serial port and process the data obtained by infrared sensors, digital compass, and accelerometer. It is able to accomplish the different challenges of HuroSot, including Penalty Kick (PK), basketball, lift- and-carry, obstacle run, robot dash, weight lifting, and marathon autonomously and effectively. The rest of this chapter is organized as follows: In Section 2, the mechanical system design of the robot TWNHR-IV is described. In Section 3, the electronic system including a vision system, a control core, and sensor systems are described. In Section 4, some simulation and experiments results of the proposed controller are described. Finally, some conclusions are made in Section 5. 2. Mechanical System Design Mechanical system design is the first step of design a humanoid robot. Human body mechanism basically consists of bones, joints, muscles, and tendons. It is impossible to replace all of the muscular-skeletal system by current mechanical and electrical components. Therefore, the primary goal of the humanoid robot mechanical system design is to develop a robot that can imitate equivalent human motions. The degrees of freedom (DOFs) configuration of TWNHR-IV is presented in Figure 1. TWNHR-IV has 26 DOFs. In this chapter, the rotational direction of each joint is defined by using the inertial coordinate system fixed on the ground as shown in Figure 1 ( Wong et al., 2008c). Fig. 1. DOFs configuration A photograph and a 3D mechanical structure of the implemented robot are shown in Figure 2 The design concepts of TWNHR-IV are light weight and compact size. The height of TWNHR-IV is 43 cm and the weight is 3.4 kg with batteries. A human-machine interface is designed to manipulate the servo motors. The details of the mechanical structure of the head, arms, waist, trunk, and legs are described as follows. Fig. 2. Photograph and mechanical structure of TWNHR-IV The head of TWNHR-IV has 2 DOFs. Figure 3 shows the 3D mechanism design of the head. The head is designed based on the concept that the head of the robot can accomplish the raw and pitch motion. Table 1 presents the head DOFs relation between human and TWNHR-IV. Human Figure TWNHR-IV Human Figure TWNHR-IV Table 1. The head DOFs relation between human and TWNHR-IV (a) Head 3D design (b) DOFs diagram Fig. 3. Head mechanism The head of TWNHR-IV has 2 degrees of freedom. Figure 4 shows the 3D mechanism design of the waist and trunk. The trunk is designed based on the concept that robot can walk and maintain its balance by using gyro to adjust the trunk motions to compensate for the robot’s walk motion. Table 2 presents the specification of the joints for the waist and trunk. RobotSoccer50 Human Figure TWNHR-IV Human Figure TWNHR-IV Table 2. The waist and trunk DOFs relation between human and TWNHR-IV (a) Waist 3D design (b) DOFs diagram Fig. 4. Waist and trunk mechanism Each arm of TWNHR-IV has 4 DOFs. Figure 5 shows the 3D mechanism design of the arms. The arms are designed based on the concept of size of the general human arms. The arms of the robot can hold an object such as a ball. Table 3 presents the specification of the joints for each arm. Human Figure TWNHR-IV Human Figure TWNHR-IV Table 3. The arms DOFs relation between human and TWNHR-IV (a) Shoulder (b) Elbow (c) Wrist (d) Arm Fig. 5. Left arm mechanism Each leg of TWNHR-IV has 7 Degrees of freedom. Figure 6 shows the 3D mechanism design of the legs. The legs are designed based on the concept that robot can accomplish the human walking motion. Table 4 presents the specification of the joints for each leg. Human Figure TWNHR-IV Human Figure TWNHR-IV Table 4. The legs DOFs relation between human and TWNHR-IV Small-sizeHumanoidSoccerRobotDesignforFIRAHuroSot 51 Human Figure TWNHR-IV Human Figure TWNHR-IV Table 2. The waist and trunk DOFs relation between human and TWNHR-IV (a) Waist 3D design (b) DOFs diagram Fig. 4. Waist and trunk mechanism Each arm of TWNHR-IV has 4 DOFs. Figure 5 shows the 3D mechanism design of the arms. The arms are designed based on the concept of size of the general human arms. The arms of the robot can hold an object such as a ball. Table 3 presents the specification of the joints for each arm. Human Figure TWNHR-IV Human Figure TWNHR-IV Table 3. The arms DOFs relation between human and TWNHR-IV (a) Shoulder (b) Elbow (c) Wrist (d) Arm Fig. 5. Left arm mechanism Each leg of TWNHR-IV has 7 Degrees of freedom. Figure 6 shows the 3D mechanism design of the legs. The legs are designed based on the concept that robot can accomplish the human walking motion. Table 4 presents the specification of the joints for each leg. Human Figure TWNHR-IV Human Figure TWNHR-IV Table 4. The legs DOFs relation between human and TWNHR-IV RobotSoccer52 (a) 3D design (b) DOFs diagram Fig. 6. Legs mechanism The head of TWNHR-IV has 2 DOFs. The head is designed based on the concept that the head of the robot can accomplish the raw and pitch motion. The trunk of TWNHR-IV has 2 DOFs. The trunk is designed based on the concept that robot can walk to adjust the trunk motions to compensate for the robot’s walk motion. Each arm of TWNHR-IV has 4 DOFs. The arms are designed based on the concept of size of the general human arms. The arms of the robot can hold an object such as a ball. Each leg of TWNHR-IV has 7 DOFs. The legs are designed based on the concept that robot can accomplish the human walking motion. The specification is shown in Table 5. Specification Height 43 cm Weight 3.4 kg Degree of Freedom & Motor Configuration DOFs Torque (kg/cm) Head 2 1.7 Thunk 2 40.8 Legs 7(x2) 37.5 Arms 4(x2) 20 Total 26 Table 5. Mechanism specification 3. Electronic System The electronic system diagram is show in Figure 7, where NIOS II is a 32-bit embedded soft- core processor implement on a FPGA chip of a development board. TWNHR-IV is using the NIOS II development board to control all of the servo motors and communicate with sensors. The DSP processor μ’nsp decides motions and gives the NIOS II development board order commands to do such as walk forward, turn right and left. The motions through the RS-232 download to the NIOS II development board. Fig. 7. System block diagram of the electronic system used for TWNHR-IV 3.1 Vision System A 16-bits DSP processor named μ’nsp is used to receive and process the image data from the CMOS image sensor via the serial transmission. The CMOS sensor is mounted on the head of the robot so that the vision information of the field can be obtained. Two main electrical parts in the vision system of the robot are a CMOS sensor and a 16-bit DSP processor. The captured image data by the CMOS sensor is transmitted to the DSP processor via a serial port. Based on the given color and size of the object, the DSP processor can process the captured image data to determine the position of the object in this image. The noise of the environmental image can be eliminated by the DSP processor. It is shown an example of color image in Figure 8. In this image, two balls are detected. The cross marks in Figure 8 (b) denote center of each color region. Based on the extracted position information, an appropriate strategy is made and transmitted to the FPGA chip via a serial transmission. [...]... in Robotics and Automation, pp 29 -34 , ISBN 0-78 03- 72 03- 4, Aug 2001 Miyazaki, F & Arimoto S (1980) A control theoretic study on dynamical biped locomotion, ASME J Dyna Syst Meas Contr., Vol.102, pp. 233 - 239 , 1980 Pauk J H & Chung H (1999) ZMP compensation by on-line trajectory generation for biped robots, IEEE International Conference on Systems, Man, and Cybernetics, Vol.4, pp 960965, ISB 0-78 03- 5 731 -0,... 5, May 2004, pp 838 -8 43, ISBN 0018-9286 Esfahani, E T & Elahinia, M H (2007) Stable walking pattern for an SMA-actuated biped, IEEE/ASME Transactions on Mechatronics, Vol 12, Issue 5, Oct 2007, pp 534 -541, ISBN 10 83- 4 435 Guan, Y.; Neo, E.S.; Yokoi, K & Tanie, K (2006) Stepping over obstacle with humanoid robot, IEEE Transaction on Robotics, Vol 22, Oct 2006, pp 958-9 73, ISBN 1552 -30 98 Hemami, H.; Barin,... Small-size Humanoid Soccer Robot Design for FIRA HuroSot 63 (a) (b) (c) (d) (e) (f) Fig 21 A vision-based humanoid soccer robot can walk along the white line autonomously in the competition of marathon 5 Conclusions A humanoid soccer robot named TWNHR-IV is presented A mechanical structure is proposed to implement a humanoid robot with 26 degrees of freedom in this chapter This robot has 2 degrees... kick 58 Robot Soccer 4.2 Basketball The basketball is one of competitions in HuroSot League of FIRA RoboWorld Cup In the competition, the robot needs to throw the ball into a red basket The robot stands in start point and then the robot need to move out of the start area When the robot move out the start area, the robot could throw the ball into the basket In the competition of basketball, the robot. .. (2008a) TWNHR-IV: Humanoid soccer robot, 5th International Conference on Computational Intelligence, Robotics and Autonomous Systems (CIRAS 2008), pp 1 73- 177, Jun 2008 Wong, C.C ; Cheng, C.T ; Huang, K.H ; Yang, Y.T.; Hu, Y.Y.; Chan, H.M & Chen, H.C (2008b) Humanoid soccer robot: TWNHR-IV, Journal of Harbin Institute of Technology, Vol.15, Sup 2, Jul 2008, pp 27 -30 , ISSN 1005-91 13 Wong, C.C.; Cheng, C.T.;... walking control robot, IEEE Int Conf on Robotics and Automation, pp 109-116, ISBN 0-8186-7728-7, Nov 1996 66 Robot Soccer Humanoid soccer player design 67 4 X Humanoid soccer player design Francisco Martín, Carlos Agüero, José María Cañas and Eduardo Perdices Rey Juan Carlos University Spain 1 Introduction The focus of robotic research continues to shift from industrial environments, in which robots must... determine 0 to 36 0 degree 60 Robot Soccer (a) Avoid obstacles (b) Walk forward Fig 16 Photographs of TWNHR-IV for the obstacle run 4.5 Robot Dash The robot dash is one of competitions in HuroSot League of FIRA RoboWorld Cup In the competition, the robot needs to go forward and backward as soon as possible The digital compass sensor is used to correct the head direction As shown in Figure 17, the robot direction... diagram of NIOS II development Small-size Humanoid Soccer Robot Design for FIRA HuroSot 55 The motions of the robot are designed on a PC, and downloaded to the RS 232 transmission module of the robot Two different data will be sent to the RS 232 transmission module, motion data and motion execution command The Data analysis module analyzes the data from the RS 232 transmission module If the command is motion... executed In that way, the robot does not waste time to stop and turn Goal direction Robot direction Fig 17 Description of the relative angle of the goal direction and the robot direction Turn Left Forward Table 8 Three motions mode Forward Turn Right Forward Small-size Humanoid Soccer Robot Design for FIRA HuroSot 61 Some pictures of TWNHR-IV playing the competition event: the robot dash are shown in... Chan, H.M.; Hu, Y.Y & Chen, H.C (2008c) Mechanical design of small-size humanoid robot: TWNHR-IV, Journal of Harbin Institute of Technology, Vol.15, Sup 2, Jul 2008, pp 31 -34 , ISSN 1005-91 13 Wong, C.C.; Huang, K.H.; Yang, Y.T.; Chan, H.M.; Hu, Y.Y.; Chen, H.C.; Hung, C.H & Lo, Y.W (2008d) Vision-based humanoid soccer robot design for marathon, 2008 CACS International Automatic Control Conference, pp . localization through sensor fusion applied to soccer robots. In Proc. Scientific Meeting of the Por- tuguese Robotics Open - ROBOTICA 20 03. Robot Soccer4 4 Martins, D. A., A. J. R. Neves, and A Image Processing, 1997, Volume 1, pp. 417–419. Robot Soccer4 6 Small-sizeHumanoid Soccer Robot DesignforFIRAHuroSot 47 Small-sizeHumanoid Soccer Robot DesignforFIRAHuroSot Ching-ChangWong,Chi-TaiCheng,Kai-HsiangHuang,Yu-TingYang,Yueh-YangHuand Hsiang-MinChan X. 0 to 36 0 degree. IR1 IR2 IR3 IR4 IR5 IR6 Robot Soccer6 0 (a) Avoid obstacles (b) Walk forward Fig. 16. Photographs of TWNHR-IV for the obstacle run 4.5 Robot Dash The robot