This paper proposes an improved rapidly-exploring random tree (RRT) algorithm, termed kinetic and dynamic-RRT (KD-RRT), which incorporates kinematic constraints and dynamic optimization to address path planning problems in dynamic environments. In response to the challenges faced by traditional RRT algorithms—namely, high randomness, poor path adaptability, and insufficient responsiveness in dynamic scenarios—the KD-RRT introduces four key mechanisms: dynamic weight target-biased sampling, multi-factor coupled dynamic step-size adjustment, kinematic constraint filtering, and incremental re-planning. These enhancements significantly improve the efficiency and quality of path planning. Experimental results demonstrate that the KD-RRT outperforms the conventional RRT algorithm in terms of path length, planning time, re-planning time, and path curvature. In particular, the proposed method exhibits greater adaptability and robustness in complex dynamic environments. This algorithm thus provides an effective solution for autonomous navigation of intelligent vehicles operating in dynamic settings.