Abstract : Based on the concept of "remote brain", this paper proposes the concept of "remote cerebellum" and designs an omnidirectional mobile robot based on Zigbee technology. The hardware implementation schemes of each functional module of the robot are given in detail and analyzed and demonstrated in depth. Keywords : Omni-directional mobile robot; remote brain; Zigbee technology; McCann wheel [align=center]Design of Omni-directional Mobile Robot Body Based on Zigbee Technology Ding Ru1, Zheng Tong1,2, Zhao Li1 (1. Automation Engineering Department, Tianjin University of Technology and Education, Tianjin 300222, China; State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin 300072, China) Abstract : This paper presents the conception of remote-cerebellum based on remote-brained conception. According to it, the Omni-directional mobile robot based on Zigbee Technology is designed. And the hardware design scheme of each function module in the robot is introduced in detail. This paper also This paper analyzes and demonstrates the realization scheme. Keywords: Omni-directional mobile robot; Remote-brained; Zigbee Technology; Mecanum wheel 1. Introduction The concept of "remote brain" [1-2] proposed by the Inoue Institute of the University of Tokyo in 1998 is an advancement in the control system architecture of mobile robots. In recent years, the application fields and application scope of mobile robots have been gradually expanding. Omni-directional mobile robots have all three degrees of freedom of planar motion, and theoretically can move at any angle and at any speed on the plane where the robot is located, so they have broad application prospects and good social benefits [3]. Based on the remote brain, this paper proposes the concept of remote cerebellum, which divides the concept of remote brain into two parts: remote brain and remote cerebellum. The remote brain is also installed outside the robot body, while the remote cerebellum is installed inside the robot body. The two and the robots use Zigbee modules to build a communication network. In this way, the remote brain controls the remote cerebellum, issues various instructions and commands to the remote cerebellum, and then the remote cerebellum controls the various functional modules inside the robot body to direct the robot to complete various tasks. The robots can also network through Zigbee to realize information exchange and resource sharing. 2. Omnidirectional Mobile Robot Body Design The humanoid robot body based on a remote brain adopts a modular design, and its structural block diagram is shown in Figure 1. It includes an intelligent module (cerebellum), a wireless communication module, a robotic arm control and drive module, a vehicle motion control and drive module, a sensor signal acquisition and processing module, and a voice module. [align=center] Figure 1. The configuration diagram of the robot body[/align] 2.1 Wireless Communication Module Since the robot's cerebellum and cerebrum are separated, and the robot has a certain range of motion, this design uses a wireless communication module to transmit information between the remote brain and the robot's cerebellum, eliminating the need for communication cables and facilitating robot movement. The remote brain processes the operator's commands and encodes the processed control instructions before sending them to the cerebellum via the wireless transmission module. After receiving the commands, the cerebellum analyzes and judges them, then transmits the decomposed instructions to the controllers of each motion module. Simultaneously, the cerebellum encodes the collected information and transmits it back to the brain via the wireless communication module. Considering the requirements for accuracy, speed and communication distance of command transmission between the remote brain and cerebellum, this design uses the CC2420EM RF module based on the ZigBee protocol from Chipcon. ZigBee is a new short-range wireless access technology. Compared with Bluetooth, it has the advantages of low cost, low power consumption, convenient networking, self-routing function and reliable data transmission [4]. The CC2420 is the first RF transceiver based on ZigBee technology launched by Chipcon. It requires very few external components, has stable performance and extremely low power consumption, and can ensure the effectiveness and reliability of short-range communication. The wireless communication device developed using this chip supports a data transmission rate of up to 250kbps and can realize multi-point to multi-point fast networking [5]. The CC2420EM module integrates the CC2420 and its required peripheral circuits. The specific interface circuit between the robot cerebellum and the RF module is shown in Figure 2. The microcontroller controls and sets the chip's operating mode via a 4-wire SPI bus (STE1, SIMO1, SOMI1, UCLK1), and implements read/write buffer data, read/write status registers, etc. The transmit/receive buffers can be set by controlling the state of the FIFO and FIFOP pin interfaces; the FIFOP pin must be connected to the microcontroller's interrupt pin. Idle channel estimation can be obtained through the CCA pin state. The timing information of transmit and receive frames can be obtained through the SFD pin state, thus determining the system's operating state; the SFD pin should be connected to the microcontroller's clock capture pin. [align=center] Fig.2 Interface of MCU and CC2420[/align] 2.2 Vehicle Motion Control and Drive Module The vehicle studied in this paper consists of a Mecanum wheel system. The Mecanum wheel system is a commonly used omnidirectional movement mechanism in mobile robots. It can move flexibly along any planar direction, and has the unique function of lateral movement and rotation in place only around its own radius [6-7]. The vehicle body adopts a four-wheel movement mechanism. The four Mecanum wheels are driven independently by four DC motors. By controlling the speed of these four motors, lateral, longitudinal, rotation around its own center, and diagonal movements can be achieved. Each Mecanum wheel consists of two parts: an active hub and multiple passive rollers evenly distributed on the outer edge of the hub in a certain inclined direction. [align=center] Fig.3. Sketch map of Mecanum wheels' configuration and speed[/align] The core of the robot's motion control is an Atmega128 chip. Each McCann wheel is independently driven by a DC motor and reducer. Each motor consists of a bridge circuit composed of four power devices (VMOS, IRLZ44N). To achieve better dynamic performance and low-speed stability, an H-type PWM speed control method is adopted, simultaneously enabling forward and reverse operation as well as braking. 2.3 Robotic Arm Control and Drive Module The robotic arm section uses two dedicated modules to control the two robotic arms respectively. The drive module controls the movement of the robotic arms, and the pose detection device detects the position and posture of the joints in real time, feeding back to the cerebellum. The cerebellum determines whether the user command has been completed. If the user command has not been completed, it continues to execute; if the user command has been completed, the robot's cerebellum notifies the remote brain via a wireless transmission module that the user command has been completed. 2.4 Sensor Signal Acquisition and Processing Module For a robot, sensors are like its eyes, ears, and nose, helping it correctly identify its environment and assisting it in successfully completing its tasks. There are many types and quantities of sensors. This paper, based on design requirements, uses bottom-detecting photoelectric sensors, edge-detecting photoelectric sensors, and ultrasonic ranging sensors. All sensors on the robot are controlled and collected by a single microcontroller, which sends information to the host computer in real time via the cerebellum. When the host computer issues a command to move the robot in a certain direction, it simultaneously sends corresponding information to the sensor module. The sensor module then queries the sensor signal in the corresponding direction and promptly feeds it back to the host computer for processing. In case of an emergency, such as when the robot is about to collide with a wall or object, the sensor module first sends an interrupt signal to the vehicle control module to stop it, and then feeds back to the host computer for processing and continues to query. The effective distance of each sensor is shown in Table 1. [align=center]Table 1. Efficiency distance of used sensor[/align] 2.5 Voice Module From a human-centered design perspective, this design includes a voice module. When encountering people or obstacles, the robot can issue an alarm through the voice module according to the requirements of the host computer or external input signals, notifying the operator or indicating the source of the obstacle, thereby avoiding accidents. The structure of the robot's voice module is shown in Figure 4. [align=center] Figure 4. Speech module frame Figure 5. Omnidirectional mobile robot[/align] 3. Experimental Research and Analysis The prototype of the omnidirectional mobile robot is shown in Figure 5. Through driving tests, the omnidirectional mobile robot can achieve smooth movements in forward, backward, left, right, left-front, right-front, left-rear, right-rear, counter-clockwise, and clockwise directions. It exhibits good linearity in straight-line movement, and the positional error is small after rotating at any angle in place. Because the robot uses a geared motor with a reduction gearbox, it can self-lock and stop on a small-angle inclined plane, and can also move slowly. The robot's movement is flexible and smooth, and its movement is not limited by the workspace, which allows for a wider range of applications. However, due to the unconsidered overall weight and dimensions, the robot's volume and weight still have room for variation. 4. Summary Based on the concept of “remote brain”, this paper proposes the concept of “remote cerebellum”. The two communicate using Zigbee technology and adopt Mecanum-type omnidirectional moving wheels and modular design method to realize the body design of the omnidirectional mobile robot, which maximizes the robot’s autonomy. Since the robot has functions such as wireless communication, omnidirectional movement, automatic obstacle avoidance and robotic gripping, it can be used in many relatively flat working situations. Moreover, based on this robot, more advanced and perfect robots can be developed. References [1] MASAYUKI INABA. Remote-brained humumanoid project [J]. Advanced Robotics, 1997, 11(6):605-620. [2] JAMES J. Dynamically-Stable Motion Planning for Humanoid Robots [J]. Autonomous Robots, 2002, 12: 105-118. [3] Xiong Rong, Zhang He, Chu Jian et al. Modeling and optimal control of four-wheeled omnidirectional mobile robot [J]. Control Theory and Applications. 2006, 23(1): 93-98 [4] Jin Chun, Jiang Xiaoyu, Luo Zuqiu. Analysis and comparison of ZigBee and Bluetooth [J]. Standards and Technology Tracking. 2004, 6: 17-20 [5] Chipcon, CC2420 2.4GHz IEEE 802.15.4/ZigBee-ready RF Transceiver Data Sheet [6] Shi Weiliang, Wang Xingsong, Jia Qian. Development of omnidirectional mobile robot based on Mecanum wheel [J]. Mechanical Engineer. 2007, 9: 18-21 [7] Lü Weiwen. Structural design of omnidirectional wheel mobile mechanism [J]. Mechanical and Electronic. 2006, 12: 63-65 Funding: Research on Multimodal Miniaturized Human-Machine Interface Technology for Perception of Space Service Robots, National High Technology Research and Development Program of China (863 Program) Project (2007AA04Z254) Author Biography: Ding Ru: (1977-) Female, from Nong'an County, Jilin Province, Lecturer, Master's degree, major research direction: Measurement and Control Technology and Instruments. Tel: 1382114603, Email: [email protected] or [email protected]