Traditional particle filtering has a large estimation error in the state of charge and Lithium-ion battery health of electric vehicle lithium batteries. For the above-mentioned problems, the lithium battery second-order resistance capacitance (RC) equivalent circuit model is established, and then, the model parameters are identified using the multi-innovation least square algorithm (MILS). Finally, an iterative unscented Kalman particle filtering algorithm with fused Rauch–Tung–Striebel Smoothing Structure (RTS-IUPF) applied to Li-ion battery state-of-charge (SOC) and state-of-health (SOH) joint estimation is proposed. The algorithm is based on the identification of battery parameters; the controller reads the sensor data and predicts the state results. RTS smoothing structure can do posterior estimation, and a significant probability density function is generated to select the optimal particle, and unscented Kalman algorithm regularized particles. The algorithm reduces the effect of the process noise covariance matrix and the measured noise covariance matrix on the filter accuracy and response time in traditional unselected Kalman filters. The algorithm proposed in the paper improves particle degradation and increases the estimation accuracy. Finally, the RTS-IUPF algorithm performs simulation analysis in Pulse current discharge condition and dynamic current condition (NEDC), respectively. The pulse current experimental results show that the mean absolute value error of UKF and particle filter (PF (number of particles N is 300)) is 1.26% and 1.24%, respectively, while the error of the RTS-IUPF is 0.748%. The root-mean-square error (RMSE) of the RTS-IUPF is reduced by 66.5% and 77.8% compared with UKF and PF. Furthermore, the error of joint estimation using this algorithm is smaller than that of single estimation. The RMSE of the RTS-IUPF joint is reduced by 27.4% compared with RTS-IUPF. The feasibility and effectiveness of the algorithm for the joint estimation of SOC and SOH of lithium batteries were verified.