Embedded System for Motion Control of an Omnidirectional Mobile Robot

ABSTRACT

In this paper, an embedded system for motion control of omni directional mobile robots is presented. An omni directional mobile robot is a type of holonomic robots. It can move simultaneously and independently in translation and rotation. The Robo Cup small-size league, a robotic soccer competition, is chosen as the research platform in this study. The first part of this research is to design and implement an embedded system that can communicate with a remote server using a wireless link, and execute received commands. Secondly, a fuzzy-tuned PI path planner and a related low-level controller are proposed to attain optimal input for driving a linear discrete dynamic model of the omnidirectional mobile robot. To fit the planning requirements and avoid slippage, velocity and acceleration filters are also employed. In particular, low-level optimal controllers, such as a linear quadratic regulator (LQR) for multiple-inputmultiple- output (MIMO) acceleration and deceleration of velocity are investigated, where an LQR controller is running on the robot with feedback from motor encoders or sensors. Simultaneously, a fuzzy adaptive PI is used as a high-level controller for position monitoring, where an appropriate vision system is used as a source of position feedback. A key contribution presented in this research is an improvement in the combined fuzzy-PI LQR controller over a traditional PI controller. Moreover, the efficiency of the proposed approach and PI controller are also discussed. Simulation and experimental evaluations are conducted with and ,without external disturbance. An optimal result to decrease the variances between the target trajectory and the actual output is delivered by the onboard regulator controller in this study. The modeling and experimental results confirm the claim that utilizing the new approach in trajectory-planning controllers results in more precise motion of four-wheeled omnidirectional mobile robots.

 

EXISTING SYSTEM:

An embedded system refers to the combination of hardware and software. There are multiple hardware and software modules required to build a robot. First of all, there are four motors connected to the omni-wheels to drive the robot. The motors can be driven by PWM or analog signals can be generated from a microcontroller using a digital to analog converter (DAC). There is a radio device used to receive commands from the team server and to send feedback. A general overview of the embedded system design and implementation is given in section III. Approximately thirty teams participated in RoboCup-SSL in 2016 from different countries around the world.

Each team must publish their Team Description Paper (TDP). However, few teams published detailed descriptions of their robot hardware and firmware. In this section, the investigation outcome of the robot hardware and firmware of referenced teams is described. For velocity evaluation, every motor has a deep quadrature encoder in CMDragons’s robot. An FPGA acts as a coordinator a the PWM generator, the quadrature decoders and the serial communication with additional onboard components. To enhance its performance, the robot also gets feedback from a gyroscope which control the robot’s movement which is specified by three different constants such as acceleration, deceleration and speed. Past teams overcame this problem by using a programmed robot behavior which finds the best set of motion profile parameters. Their algorithm tried to solve this trade-off between the overall robot velocity and motion instability.

This algorithm runs on an FPGA. Tigers Mannheim’s processor can control all motors, perform wireless communication, and read the encoders. In order to fulfill the requirements of hardware interrupts, they used the FreeRTOS real-time operating system . To control the robot kinematics such as angular velocity and position, four brush-less motors are used team’s robot. These motors are controlled by the microprocessor core (RISC-32 architecture) in SKUBA robot . To control the driving sequence, a digital sequencer FPGA module is used. Two problems related to the driving mechanism can destroy the robot: over-current and motor dead-time. In the motor data-sheet, the dead-time value is given. Dead-time is a blanking time period (upper & lower transistors in off-state simultaneously) of half-bridge power stage. The dead-time protection is built in the FPGA to prevent over current whereas a fully Integrated hall-effectbased linear current sensor IC (ACS712) calculates the motor driving current to protect circuits. If the current exceeds the limit, the firmware will keep the PWM signals less than the motor’s maximum range. Additionally, robotic applications must adapt to dynamic environments. For plane trajectory following, speed and position accuracy of an omnidirectional robot are key subjects for extensive studies . The development of the dynamic and kinematic equations with consistent control was the main purpose of numerous studies.

The dynamics and kinematics of omnidirectional mobile robots are shown in a few articles. Robot dynamic model have been developed after linearized systems . More specifically, two autonomous PID controllers are built for controlling direction and position differently based on an easy linear model. However, a nonlinear relationship a the translational and rotational velocities has not yet been investigated. The trajectory follower robot where a geometric path is given and the robot tries to follow the exact same path using feedbacks, by Chen . Kalmar  details creating a symmetrical path and using feedback to track the trajectory. Also, the dynamic capabilities of the mobile robots are measured and applied to swiftly embryonic environments . Moreover, a control approach and optimized maneuver planing for robot position control without considering orientation has been established already. A continuous and nonlinear model has also been demonstrated, where an applied technique to filter speed and dynamics to achieve a slippage-free (i.e. without slippage of the wheels) drive.

 

PROPOSED SYSTEM :

In this paper, an embedded system for motion control of omni directional mobile robots is presented. An omni directional mobile robot is a type of holonomic robots. It can move simultaneously and independently in translation and rotation. The Robo Cup small-size league, a robotic soccer competition, is chosen as the research platform in this study. To fit the planning requirements and avoid slippage, velocity and acceleration filters are also employed. In particular, low-level optimal controllers, such as a linear quadratic regulator (LQR) for multiple-inputmultiple- output (MIMO) acceleration and deceleration of velocity are investigated, where an LQR controller is running on the robot with feedback from motor encoders or sensors. Simultaneously, a fuzzy adaptive PI is used as a high-level controller for position monitoring, where an appropriate vision system is used as a source of position feedback. In this system is simple and easy-to-use embedded system is developed for controlling the motion of four wheeled omni directional mobile robots using two closed loop structure control systems that can successfully read commands from the team server and execute received commands to drive the robot accurately and efficiently in an x-y plane. The outer loop controller, a fuzzy adaptive PI, calculates the accurate target position and minimizes tracking errors using vision feedback. Concurrently, an inner loop controller, a discrete time linear quadratic regulator is employed on the robot to obtain the  optimal input for the motor drivers to help the robot to run with an exact wheel speed in an energy efficient way.

 

CONCLUSION :

 

In this Paper, a simple and easy-to-use embedded system is developed for controlling the motion of four wheeled omni directional mobile robots using two closed loop structure control systems that can successfully read commands from the team server and execute received commands to drive the robot accurately and efficiently in an x-y plane. The outer loop controller, a fuzzy adaptive PI, calculates the accurate target position and minimizes tracking errors using vision feedback. Concurrently, an inner loop controller, a discrete time linear quadratic regulator is employed on the robot to obtain the  optimal input for the motor drivers to help the robot to run with an exact wheel speed in an energy efficient way. An essential contribution illustrated in this research is an improved performance in the combination of fuzzy tuned PI with LQR controller over classic PI controller. Furthermore, a conducted  simulation and experiment with and without external load yield.