1: \begin{abstract}

2:         We consider Riemannian inequality-constrained optimization problems and propose a Riemannian primal-dual interior point trust region method (\RIPTRM{}) for solving them.

3:         We prove its global convergence to an approximate Karush-Kuhn-Tucker point and a second-order stationary point.

4:         We also establish the local near-quadratic convergence.

5:         To the best of our knowledge, this is the first algorithm that incorporates the trust region strategy and has the second-order convergence property for optimization problems on Riemannian manifolds with nonlinear inequality constraints.

6:         It is also the first Riemannian interior point method that possesses both global and local convergence properties.

7:         We conduct numerical experiments in which we introduce a truncated conjugate gradient method and an eigenvalue-based subsolver for \RIPTRM{} to approximately and exactly solve the trust region subproblems, respectively.

8:         Empirical results show that \RIPTRM{}s find solutions with higher accuracy compared to an existing Riemannian interior point method and other algorithms.

9:         Additionally, we observe that \RIPTRM{} with the exact search direction shows significantly promising performance in an instance where the Hessian of the Lagrangian has a large negative eigenvalue.

10:     \end{abstract}

11: