Research on Motion Control of Berry Picking Robot Based on ARM

Abstract: A motion control system for a berry-picking robot is constructed based on an ARM microprocessor. The hardware and software of the system are designed separately, and a feedback-enabled closed-loop control scheme is proposed. This control system has advantages such as modularity, easy expansion, portability, small hardware size, low power consumption, strong real-time performance, and high reliability.

Keywords: ARM; harvesting robot; motion control

Abstract： Movement control system of berry picking manipulator is established, with ARMmicroprocessor as the core. The software and hardware of the system are designed respectively and a closed-loop control scheme capable of feeding back is presented. The control system features modularization, easy extension, transportability, small volume of hardware, low powerconsumption, strong real-time and high reliability.

Key words: ARM; pickingmanipulator; movement control

With the rapid development of computer and automatic control technologies, agricultural machinery is entering a period of high automation and intelligence. The application of berry-harvesting robots can improve labor productivity and product quality, improve working conditions, and solve problems such as labor shortages. A berry-harvesting robot mainly consists of a robotic arm and end effector, a vision and decision-making system, and a control system. This paper will explain how to use an ARM microprocessor to achieve motion control of the berry-harvesting robot arm. This control system uses an ARM (Advanced RISC Machine) microprocessor, which has strong versatility compared to microcontrollers and DSPs, and is widely used in various fields due to its high speed, high cost-effectiveness, and low power consumption.

1. Control System Functional Requirements

The main actuator of the berry-harvesting robot—the robotic hand—is divided into two parts: the arm and the wrist. How the robotic hand avoids obstacles and accurately reaches the target fruit is determined by its degrees of freedom (DOF). Typically, the robotic hand's position and range of motion in space mainly depend on the ODF of the arm. To enable the robotic hand to reach any specified position in space, the ODF should have at least three DDFs. The wrist's DDFs are mainly used to adjust the end effector's posture in space. Theoretically, to allow the end effector to achieve any desired posture in space, the wrist should also have three DDFs. A simplified schematic diagram is shown in Figure 1.

The task of the control system is to receive instructions from the host computer and drive the motors corresponding to each degree of freedom, so that the harvesting robot can reach the designated position to perform operations.

2 Control System Hardware Design

Robot control systems generally need to meet the following basic requirements: ① Miniaturization, lightweight design, and modularity of the control system; ② Real-time performance of the control system; ③ System stability and openness.

Therefore, this control system is designed to consist of three parts: a main control module, a drive module, and a feedback module, as shown in Figure 2.

2.1 Main Control Module Design

To meet the robot's control requirements while also considering the robot's sensitive requirements regarding controller size, weight, and power consumption, the main control module uses Samsung's S3C2410, a low-power, 16/32-bit, high-performance RISC microprocessor based on the ARM920T core (suitable for real-time environments), with a clock speed of 266MHz. The operating system chosen is the open-source, highly reliable real-time, multi-tasking arm-Linux kernel, specifically designed for ARM.

For a microprocessor to function properly, its peripherals must be expanded. Figure 3 shows the hardware system schematic of the entire ARM main control module.

The power supply module provides a stable and clean DC power supply for the entire module. The JTAG debugging unit is used for online debugging of the program. The serial communication module enables communication between the main control module and the host computer via the RS232 communication standard. To enable the system to run larger programs (such as the Linux kernel and file system), a 32MB SDRAM memory chip is added around the microprocessor. A 16MB FLASH chip is also added as a storage device for programs and data to ensure that programs and data are not lost in the event of power failure.

2.2 Driver Module Design

This robot system uses DC motor control, and the motor driver chip selected is STMicroelectronics' L298. The L298 is a high-voltage, high-current motor driver chip manufactured by STMicroelectronics. Its main features include high operating voltage, large output current (up to 3A instantaneous peak current and 2A continuous operating current), two H-bridge high-voltage, high-current full-bridge drivers (capable of driving DC motors, stepper motors, relays, and inductive loads such as coils), standard TTL logic level signal control, two enable control terminals to allow or disable device operation without being affected by input signals, a logic power input terminal to allow the internal logic circuitry to operate at low voltage, and the ability to connect an external sensing resistor to feed back changes to the control circuit.

2.3 Feedback Module Design

The system uses photoelectric encoders to measure the speed of the motor. Photoelectric encoders are characterized by low inertia, low noise, high resolution, and high precision, making them suitable for controlling DC motors. The pulse signals generated by the encoder are transformed to obtain the motor speed, which is used for speed feedback to form a speed closed loop. At the same time, the pulse signals can be counted and calculated to obtain the position and speed of the harvesting robot.

3 Control System Software Design

In a control system, control tasks are implemented by an application program, and the quality of the application program design directly determines the overall control quality and efficiency. To facilitate system debugging and functional expansion, the control system software also adopts a modular design. The main program is divided into two main parts: an initialization module and a runtime module.

3.1 Initialization Module

The initialization module needs to perform the following tasks: defining the exception vector table, initializing the stack, initializing system variables, initializing the interrupt system, initializing I/O, and initializing peripheral components.

3.2 Running Module

The operation module operates via interrupts and includes a speed measurement module, a PID control module, and a PWM wave output module. The main workflow is as follows: First, it checks for new speed commands. If so, it determines the speed magnitude and direction and converts them into a standardized form specified in the program for PID control. Then, it checks for new sensor feedback values. If so, it calculates the motor speed and executes the motor's PID control program. Next, it calls the motor driver program, changes the PWM duty cycle, and outputs a PWM wave to achieve motor speed control. The speed measurement module samples the photoelectric encoder pulses at regular intervals to obtain speed feedback values. The main flow of the operation module is shown in Figure 4.

4. Conclusion

ARM-based motion control systems are an important component of berry-harvesting robot systems. ARM microprocessors offer high performance, low power consumption, small size, and good portability. Motion control systems based on ARM microprocessors can achieve portability through hardware platform improvements and upgrades, software algorithm regeneration, and modularization, thus possessing significant application value.

References:

[1] Zhou Ligong. ARM Microcontroller Fundamentals and Practice [M]. Beijing: Beijing University of Aeronautics and Astronautics Press, 2003.

[2] Qiu Bo, Li Qunzhi, Wang Huijuan. Design and implementation of master-slave gripping control system for medical robot based on ARM microcontroller [J]. Mechatronics, 2006, (6).

[3] Wang Xiaoming. Microcontroller Control of Electric Motors [M]. Beijing: Beijing University of Aeronautics and Astronautics Press, 2002.

[4] Hu Xiaojuan. Research on motion control of mobile robots based on ARM [D]. Nanjing: Nanjing University of Science and Technology, 2007.

[5] USER'SMANUALS3C2410X32-Bit RISC Microprocessor Revision0.1.Samsuang electronics［Z］