Decision making and motion planning algorithms mainly affect the safety and handling stability of autonomous vehicles on highway curve. For the issue of insufficient safety and driving efficiency in decision-making of highway lane change, a driving dissatisfaction decision algorithm based on the relative driving dissatisfaction of the ego-vehicle to the preceding vehicle was proposed. In order to improve the real-time performance of the planning algorithm, a path-speed decoupling framework was adopted for lane change trajectory planning. For path planning, the quintic polynomial curve was chosen and four path evaluation indicators that considered safety, comfort, and efficiency were used to achieve optimal path planning. For speed planning, the smooth speed curve was obtained by combining dynamic programming and quadratic programming optimization. The simulation results show that the lane change decision model based on driving dissatisfaction can choose a more efficient and safer driving mode. In typical lane changing scenarios, the maximum centroid sideslip angle and maximum yaw rate of the ego-vehicle are both small, which shows the lane change trajectory planning algorithm can ensure the safety and handling stability of ego-vehicle in the process of lane change.