Abstract: The control of a bomb disposal robot can be divided into the walking control of the vehicle and the motion control of the manipulator. The walking control of the vehicle requires "path planning" to achieve obstacle avoidance and approach the target object (suspected explosive device), while the motion control of the manipulator requires "trajectory planning" to avoid obstacles and achieve collision avoidance, successfully grasping and transporting the target object. This robot control system is an open system, realizing intelligence and networking. The remote management computer can display the robot's gripper, target object, and obstacles on site. Keywords: Explosive-removal robot; control system; network Abstract: A robot used to eliminate explosive objects needs to control its walking and the motion of its arm. Walking control, also known as "track-out," allows the robot to shun obstacles and catch and remove explosive objects. The control system is an open, network-connected, and intelligent system. In a remote administration room, the robot's arm, the explosive object, and obstacles can be viewed using a CRT computer. Key words: Explosive-removal robot; control system; network Introduction Explosive-removal robots belong to a type of handling robot. Many handling robots mimic human movements to assist or partially replace humans in performing heavy, dangerous, and repetitive tasks. Explosive-removal robots, mimicking human walking and object-grabbing movements, can be used to dispose of explosives at the scene. The mechanical parts of an explosive-removal robot include a walking cart and a robotic arm. 1. Bomb Disposal Robot Control System Structure The bomb disposal robot control system no longer uses the PLC control method previously employed by robots. Instead, it is based on an embedded industrial computer and utilizes an open component library based on the Linux platform, achieving extremely high processing speeds, reaching microsecond levels. The bomb disposal robot control system consists of three parts: the sensor section, the robot body, and the operation console. The sensor section includes a wide-access, multi-information fusion unit. This unit, based on an ARM9 system, consists of a 12-bit or higher precision A/D converter, 32 I/O channels, and an RS232 communication interface, capable of fusing multiple sensor signals. The robot body includes an embedded control computer and servo motors for each joint, capable of driving each axis (with compensation), allocating axis motion, and processing and controlling various state variables. The operation console includes a main control PC, instruction decoder, image decoder, and wireless receiver/transmitter. Image and obstacle information obtained from the CCD camera and ultrasonic sensors are wirelessly transmitted to the main control PC. The main control PC uses image recognition and analysis technology to obtain the shape and position information of the target object, and simultaneously displays the three-dimensional image information of the target object and obstacles on the computer screen. After the target object is manually indicated, the main control PC calculates using artificial intelligence (pattern recognition, path planning, trajectory collision avoidance, etc.) and transmits control commands wirelessly to the embedded control computer in the robot body, automatically controlling the robot's movement. The basic structure is shown in Figure 1: [align=center] Figure 1 Basic Structure of the Bomb Disposal Robot Control System[/align] 2. Vehicle Movement Control Manually, using control levers (or buttons) and guided by the vehicle's image (visual signal), the robot moves forward, backward, turns left, turns right, and rotates in place until the robotic arm's gripper can grasp the suspected explosive device. The robot's balance is achieved using a dynamic algorithm, which uses various data to control the robot arm's spatial posture to maintain overall balance and prevent tipping. For example, when the vehicle is moving upwards on a slope, the robot arm moves forward to maintain balance. The vehicle needs to avoid obstacles during its movement. By acquiring obstacle size and environmental information through environmental sensors such as CCD cameras and ultrasonic sensors, the robot can avoid obstacles, which is called path planning (as shown in Figure 2). [align=center] Figure 2 Schematic diagram of path planning for bomb disposal robot[/align] 3. Motion control of the robotic arm The motion control of the robotic arm (as shown in Figure 3) is a multi-level control system: the top-level AI layer is the artificial intelligence layer. It uses CCD cameras and ultrasonic sensors to obtain the outline and distance of obstacles to avoid collisions, which is "trajectory planning" (as shown in Figure 4). [align=center] Figure 3 Multi-level motion control system of robotic arm Figure 4 Schematic diagram of trajectory planning of robotic arm[/align] The control layer in the middle is the most critical part of the robotic arm's grasping motion control. It is actually an interpolator. The motion control of the robotic arm can be divided into posture control and hand position control. The control of the waist, upper arm, forearm, and gripper is achieved by the main control computer solving the kinematic displacement problem of the robotic arm in reverse motion. This determines the positions of the waist, upper arm, forearm, and gripper, and then sends position commands to each joint to the embedded system control computer of those joints, completing position control with a speed feedback loop. The movement speed of the four joints is determined by the robotic arm's trajectory planning, thus avoiding serpentine movements. Thanks to tactile sensors, the gripper can be controlled automatically or with fine, "soft" control guided by image signals. When grasping a suspected explosive device, the gripper's clamping force is controlled to be approximately greater than the weight of the explosive device when it falls, preventing excessive force that could damage it, thus completing the expert-specified action. At the lowest level, the control-level position controller and speed regulator both employ mature PID (including P, PI, PD) control. When PID control is configured with appropriate parameters, it exhibits robustness, simplicity, and practicality. When PID cannot achieve effective control, it can be improved or combined with other methods, such as feedforward control. Motion control includes the motion control of each joint of the robotic arm and walking motion. Preliminary statistics show there are 13 control loops, as shown in Figure 5. [align=center]Figure 5 Automatic Control Principle Diagram[/align] The embedded control computer sets the position and speed of each servo mechanism based on the position and speed signals transmitted from the host computer, and controls the servo motors according to control laws (such as PID) to ensure the actual position tracks or moves to the target position at the expected speed. The robot's hand-eye coordination system is one of the biggest highlights of this system. Because the interpolation algorithm requires knowledge of the target object and the robot's base coordinates (usually the robot's waist joint base), the relative spatial coordinates need to be solved. However, the target object in this system is in an unknown environment. These relative coordinates must be solved using the hand-eye coordination system. The bomb disposal robot's robotic arm is equipped with a binocular system. Using the binocular system and the target object, a "moment measurement triangle" can be formed, as shown in Figure 6. The distance obtained by the moment measurement triangle has a large error, requiring the use of a laser rangefinder's "spot" to assist in moment measurement, thereby obtaining the spatial coordinates of the object. As the robot's binoculars approach the target object, the relative coordinate accuracy between the target object and the base coordinates improves. After several binocular approaches, the robotic arm finally accurately grasps the target object. [align=center]Figure 6. Schematic diagram of the moment measurement triangle[/align] When the robot's automatic grasping fails, the operator can complete the grasping using the manual system. This is a necessary function for any product after automatic failure. The image information obtained by the robot's binoculars is processed and displayed on the console monitor to form a monitoring image, allowing the operator to click on suspicious objects among multiple objects with the mouse. 4. Communication System of the Control System The robot communicates wirelessly with the main control computer of the on-site control console via a serial port. The on-site main control computer and the remote monitoring center use CDMA communication technology for wireless image transmission. The status of the target object and the robot can be observed from any point in the communication system. In laboratory and industrial applications, due to channel cost limitations, serial ports are often the preferred data transmission channel between computers and external serial devices. Furthermore, because serial communication is convenient and easy to implement, many devices and computers can control and monitor peripherals via serial ports. Serial communication is increasingly becoming a crucial means for computers and peripherals to communicate and acquire monitoring data collected by peripherals. To overcome the limitations of wired communication, and considering the characteristics of robots—their strong self-planning, self-organizing, and adaptive capabilities, and their uncertain environments—wireless communication is the ideal way to achieve communication between robots and computers. Therefore, using wireless data transmission modules for data signal transmission is a vital component of bomb disposal robots. Wireless data transmission modules are small, easy to use, and can remotely control robots, effectively protecting operator safety. CDMA (Code Division Multiple Access) is a carrier modulation and multiple access technology based on extended communication systems. CDMA has many advantages, some of which are inherent to the extended system, and others are brought about by software switching and power control. The CDMA communication network is composed of spread spectrum network, multiple access, cellular network and frequency reuse technologies. It contains a three-dimensional signal processing of frequency domain, time domain and code domain. Therefore, it has good anti-interference performance, anti-multipath fading and high security performance. 5 Conclusion and prospect The innovation of this paper is that the bomb disposal robot control system is an open control system. The control system successfully combines binocular vision ranging and intelligent control algorithm to realize the intelligent and networked functions of the robot. The applicability of this design is very wide. As long as it is appropriately modified according to the needs, a control system applicable to other functional robots can be designed. It has high reference value. Through the research and development of this system, the automatic control level of the bomb disposal robot will be improved, the operability and reliability of the bomb disposal robot will be improved, and it will be of great help to ensure public safety and improve bomb disposal efficiency. References: [1] Xu Yongxi, Jiang Liangzhong. Bomb disposal robot control system based on MatlabRTW. Microcomputer Information, Article No.: 1008-0570 (2006) 02-2-0218-03 [2] Zhou Xuecai, Li Weiping, Li Qiang. General control system for open robots [J]. Robot, 1998 (01). [3] Fu Jingxun. Robotics - Control, Sensor Technology, Vision and Intelligence. Beijing: China Science and Technology Press, 2000 (04). Design of human-powered pneumatic control system [J]. Hydraulics and Pneumatics, 2000 (04). [4] Zhang Jialiang, Lü Tiansheng, Yao Xianggen, Qu Yiyi. Cable machine