Abstract: During motion, vibrations are easily generated at corners due to the flexibility of the end actuator in Cartesian coordinate dispensing robots, leading to contour errors in the dispensing process. To reduce this error, we propose a trajectory-smoothing method that accounts for the dynamic characteristics of the end effector in robotics. First, the trajectory to be smoothed is input into the dynamic simulation model of the dispensing end effector of the robot, and the actual trajectory curve is generated. Then, based on the contour error between the actual trajectory curve and the target trajectory curve, the path curve is optimized and adjusted. Finally, the path is fitted to a C²-continuous B-spline curve, a new speed planning strategy is generated, and the entire trajectory smoothing process is completed. Experimental results demonstrate that the proposed algorithm ensures continuous displacement, velocity, and acceleration during motion, substantially improving fluid dispensing robot trajectory smoothness. Additionally, it dramatically reduces contour errors at corners, achieving an average contour error reduction of 39.65% and a root mean square contour error reduction of 30.74%, thereby enhancing dispensing accuracy.