Efficient path planning and following for non-holonomic robot

![CDATA[Embedded robotic control usually faces the dilemma of low computing-power or strict real-time performance requirement. The processing speed of microprocessor is highly constrained from the clock cycle and architectures of data path, compiler, and algorithms. In this paper, we propose an efficient algorithm to resolve difficulties in the common form of Reeds-Shepp turning curves, which can be deployed for non-holonomic robot path planning and following. The salience of our algorithm is on removing the need for the use of trigonometric, logarithm, and sophisticated matrix functions. The robot's turning schemes can then be easily implemented on a microprocessor or field-programmable-gate-array (FPGA) chip with minimum computing effort. The proposed algorithm is promising for miniature robot maneuvering controls in the future.]]