Exact hess template

acados/ocp_nlp/ocp_nlp_constraints_bgh.c

@@ -1305,8 +1305,8 @@ void ocp_nlp_constraints_bgh_bounds_update(void *config_, void *dims_, void *mod
     ocp_nlp_constraints_bgh_cast_workspace(config_, dims, opts_, work_);
 
     // extract dims
-    int nx = dims->nx;
-    int nu = dims->nu;
+    // int nx = dims->nx;
+    // int nu = dims->nu;
     int nb = dims->nb;
     int ng = dims->ng;
     int nh = dims->nh;

acados/ocp_nlp/ocp_nlp_constraints_bgp.c

@@ -1292,8 +1292,8 @@ void ocp_nlp_constraints_bgp_bounds_update(void *config_, void *dims_, void *mod
     ocp_nlp_constraints_bgp_cast_workspace(config_, dims, opts_, work_);
 
     // extract dims
-    int nx = dims->nx;
-    int nu = dims->nu;
+    // int nx = dims->nx;
+    // int nu = dims->nu;
     int nb = dims->nb;
     int ng = dims->ng;
     int nphi = dims->nphi;

acados/ocp_nlp/ocp_nlp_sqp.c

@@ -595,7 +595,7 @@ int ocp_nlp_sqp(void *config_, void *dims_, void *nlp_in_, void *nlp_out_,
         {
             printf("\n------- sqp iter %d (max_iter %d) --------\n", 
                 sqp_iter, opts->max_iter);
-            if (opts->print_level > sqp_iter)
+            if (opts->print_level > sqp_iter + 1)
                 print_ocp_qp_in(nlp_mem->qp_in);
         }
 
@@ -736,10 +736,10 @@ int ocp_nlp_sqp(void *config_, void *dims_, void *nlp_in_, void *nlp_out_,
             omp_set_num_threads(num_threads_bkp);
 #endif
 
-            if (opts->print_level > 0)
+            if (opts->print_level > 1)
             {
                 printf("\n Failed to solve the following QP:\n");
-                if (opts->print_level > sqp_iter)
+                if (opts->print_level > sqp_iter + 1)
                     print_ocp_qp_in(nlp_mem->qp_in);
             }
 
@@ -797,10 +797,6 @@ int ocp_nlp_sqp(void *config_, void *dims_, void *nlp_in_, void *nlp_out_,
 
     if (opts->print_level > 0)
     {
-        // print_ocp_qp_in(nlp_mem->qp_in);
-
-        printf("\n ocp_nlp_sqp: maximum iterations reached, last QP input above.\n");
-
         printf("Residuals: stat: %e, eq: %e, ineq: %e, comp: %e.\n", mem->nlp_res->inf_norm_res_g,
             mem->nlp_res->inf_norm_res_b, mem->nlp_res->inf_norm_res_d, mem->nlp_res->inf_norm_res_m );
     }
@@ -981,6 +977,17 @@ void ocp_nlp_sqp_get(void *config_, void *dims_, void *mem_, const char *field,
         double **value = return_value_;
         *value = mem->stat;
     }
+    else if (!strcmp("statistics", field))
+    {
+        int n_row = mem->stat_m<mem->sqp_iter+1 ? mem->stat_m : mem->sqp_iter+1;
+        double *value = return_value_;
+        for (int ii=0; ii<n_row; ii++)
+        {
+            value[ii+0] = ii;
+            for (int jj=0; jj<mem->stat_n; jj++)
+                value[ii+(jj+1)*n_row] = mem->stat[jj+ii*mem->stat_n];
+        }
+    }
     else if (!strcmp("stat_m", field))
     {
         int *value = return_value_;

acados/ocp_nlp/ocp_nlp_sqp_rti.c

@@ -503,7 +503,7 @@ int ocp_nlp_sqp_rti(void *config_, void *dims_, void *nlp_in_, void *nlp_out_,
 
 }
 
-int ocp_nlp_sqp_rti_preparation_step(void *config_, void *dims_,
+void ocp_nlp_sqp_rti_preparation_step(void *config_, void *dims_,
     void *nlp_in_, void *nlp_out_, void *opts_, void *mem_, void *work_)
 {
     acados_timer timer1;
@@ -516,13 +516,12 @@ int ocp_nlp_sqp_rti_preparation_step(void *config_, void *dims_,
     ocp_nlp_in *nlp_in = nlp_in_;
     ocp_nlp_out *nlp_out = nlp_out_;
     ocp_nlp_memory *nlp_mem = mem->nlp_mem;
-    ocp_qp_xcond_solver_config *qp_solver = config->qp_solver;
+    // ocp_qp_xcond_solver_config *qp_solver = config->qp_solver;
 
     ocp_nlp_sqp_rti_workspace *work = work_;
     ocp_nlp_sqp_rti_cast_workspace(config, dims, opts, mem, work);
     ocp_nlp_workspace *nlp_work = work->nlp_work;
 
-	double tmp_time;
     mem->time_lin = 0.0;
     mem->time_reg = 0.0;
 
@@ -701,16 +700,10 @@ int ocp_nlp_sqp_rti_preparation_step(void *config_, void *dims_,
         nlp_out, nlp_opts, nlp_mem, nlp_work);
 
     mem->time_lin += acados_toc(&timer1);
-
-    // regularize Hessian
-    acados_tic(&timer1);
-    config->regularize->regularize_hessian(config->regularize,
-        dims->regularize, opts->nlp_opts->regularize, nlp_mem->regularize_mem);
-
-    mem->time_reg += acados_toc(&timer1);
 }
 
-int ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
+
+void ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
     void *nlp_in_, void *nlp_out_, void *opts_, void *mem_, void *work_)
 {
     acados_timer timer1;
@@ -743,11 +736,16 @@ int ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
     ocp_nlp_approximate_qp_vectors_sqp(config, dims, nlp_in,
         nlp_out, nlp_opts, nlp_mem, nlp_work);
 
+    // regularize Hessian
+    acados_tic(&timer1);
+    config->regularize->regularize_hessian(config->regularize,
+        dims->regularize, opts->nlp_opts->regularize, nlp_mem->regularize_mem);
+    mem->time_reg += acados_toc(&timer1);
+
     if (opts->print_level > 0) {
         printf("\n------- qp_in --------\n");
         print_ocp_qp_in(nlp_mem->qp_in);
     }
-    // exit(1);
 
     if (!opts->warm_start_first_qp)
     {
@@ -812,7 +810,7 @@ int ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
         omp_set_num_threads(num_threads_bkp);
 #endif
         mem->status = ACADOS_QP_FAILURE;
-        return mem->status;
+        return;
     }
 
     ocp_nlp_update_variables_sqp(config, dims, nlp_in,
@@ -829,7 +827,7 @@ int ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
     omp_set_num_threads(num_threads_bkp);
 #endif
     mem->status = ACADOS_SUCCESS;
-    return mem->status;
+
 }
 
 
@@ -1016,6 +1014,17 @@ void ocp_nlp_sqp_rti_get(void *config_, void *dims_, void *mem_,
         double **value = return_value_;
         *value = mem->stat;
     }
+    else if (!strcmp("statistics", field))
+    {
+        int n_row = 2;
+        double *value = return_value_;
+        for (int ii=0; ii<n_row; ii++)
+        {
+            value[ii+0] = ii;
+            for (int jj=0; jj<mem->stat_n; jj++)
+                value[ii+(jj+1)*n_row] = mem->stat[jj+ii*mem->stat_n];
+        }
+    }
     else if (!strcmp("stat_m", field))
     {
         int *value = return_value_;

acados/ocp_nlp/ocp_nlp_sqp_rti.h

@@ -156,10 +156,10 @@ int ocp_nlp_sqp_rti_workspace_calculate_size(void *config_, void *dims_, void *o
 int ocp_nlp_sqp_rti(void *config_, void *dims_, void *nlp_in_, void *nlp_out_,
     void *opts_, void *mem_, void *work_);
 //
-int ocp_nlp_sqp_rti_preparation_step(void *config_, void *dims_,
+void ocp_nlp_sqp_rti_preparation_step(void *config_, void *dims_,
     void *nlp_in_, void *nlp_out_, void *opts, void *mem_, void *work_);
 //
-int ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
+void ocp_nlp_sqp_rti_feedback_step(void *config_, void *dims_,
     void *nlp_in_, void *nlp_out_, void *opts_, void *mem_, void *work_);
 //
 void ocp_nlp_sqp_rti_config_initialize_default(void *config_);

acados/utils/external_function_generic.c

@@ -734,6 +734,7 @@ void external_function_casadi_wrapper(void *self, ext_fun_arg_t *type_in, void *
                 d_cvt_dvec_to_casadi(in[ii], (double *) fun->args[ii],
                                      (int *) fun->casadi_sparsity_in(ii));
                 break;
+
             case COLMAJ_ARGS:
                 d_cvt_colmaj_args_to_casadi(in[ii], (double *) fun->args[ii],
                                             (int *) fun->casadi_sparsity_in(ii));
@@ -781,6 +782,7 @@ void external_function_casadi_wrapper(void *self, ext_fun_arg_t *type_in, void *
                 d_cvt_casadi_to_dvec((double *) fun->res[ii], (int *) fun->casadi_sparsity_out(ii),
                                      out[ii]);
                 break;
+
             case COLMAJ_ARGS:
                 d_cvt_casadi_to_colmaj_args((double *) fun->res[ii],
                                             (int *) fun->casadi_sparsity_out(ii), out[ii]);

docs/interfaces/index.md

@@ -49,10 +49,8 @@ Note: This part of the MATLAB/Octave interface does not yet support all features
 ## Python (templates)
 
 `acados_template` is a Python package that can be used to specify optimal control problems from Python and to generate self-contained C code that uses the acados solvers to solve them.
-In comparison with the MATLAB interface for rapid prototyping (see above), it supports less features, but it allows the user to generate a self-contained C library
-that can be easily deployed on an embedded system.
-
-The framework is based on templated C files which are rendered from Python using the templating engine `Tera`.
+The pip package is based on templated code (C files, Header files and Makefiles), which are rendered from Python using the templating engine `Tera`.
+The genereated C code can be compiled into a self-contained C library that can be deployed on an embedded system.
 
 ### Optimal Control Problem description
 The Python interface relies on the same problem formulation as the MATLAB interface [see here](https://github.com/acados/acados/blob/master/docs/problem_formulation/problem_formulation_ocp_mex.pdf).
@@ -69,17 +67,25 @@ make install -j4
 ```
 pip3 install <acados_root>/interfaces/acados_template
 ```
+Note: If you are working with a virtual Python environment, use the `pip` corresponding to this Python environment instead of `pip3`.
+
+3. Add the path to the compiled shared libraries `libacados.so, libblasfeo.so, libhpipm.so` to `LD_LIBRARY_PATH` (default path is `<acados_root/lib>`) by running:
+```bash
+export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:"<acados_root>/lib"
+```
+Tipp: you can add this line to your `.bashrc`/`.zshrc`.
+
+4. Run a Python example to check that everything works.
+We suggest to get started with the example
+`<acados_root>/examples/acados_python/getting_started/minimal_example_ocp.py`.
 
-(Notice that, you might need to use `pip` instead, if you run, for example, from within a Python virtual environment)
-You should now be able to import it as a Python module and use it as shown in the examples in `<acados_root>/examples/acados_template/python/<example_name>/generate_c_code.py`.
+5. Optional: Can be done automatically through the interface:
+In order to be able to successfully render C code templates, you need to download the `t_renderer` binaries for your platform from <https://github.com/acados/tera_renderer/releases/> and place them in `<acados_root>/bin` (please strip the version and platform from the binaries (e.g.`t_renderer-v0.0.34 -> t_renderer`).
+Notice that you might need to make `t_renderer` executable.
+Run `export ACADOS_SOURCE_DIR=<acados_root>` such that the location of acados will be known to the Python package at run time.
 
-In order to be able to successfully render C code templates,
-you need to download the `t_renderer` binaries for your platform
-from <https://github.com/acados/tera_renderer/releases/> and
-place them in `<acados_root>/bin` (please strip the version and platform from the binaries (e.g.
-`t_renderer-v0.0.34 -> t_renderer`). Notice that you might need to make `t_renderer` executable. Run
-`export ACADOS_SOURCE_DIR=<acados_root>` such that the location of acados will be known to the Python
-package at run time. Additionally, you will have to make sure that the environment variable `LD_LIBRARY_PATH` contains the path to `libacados.so` (default path is `<acados_root/lib>`). Notice that, if you want to run the examples from a location that differs from '<acados_root>/interfaces/acados_template' or you want the generated Makefile to refer to a specific path (e.g. when cross-compiling or compiling from a location different from the one where you generate the C code), you will have to adapt 'ocp.acados_include_path' and 'ocp.acados_lib_path' accordingly in the generating Python code.
+6. Optional: Set `acados_lib_path`, `acados_include_path`.
+If you want the generated Makefile to refer to a specific path (e.g. when cross-compiling or compiling from a location different from the one where you generate the C code), you will have to set these paths accordingly in the generating Python code.
 
 For more information contact `@zanellia`.
 

examples/acados_matlab_octave/getting_started/check_acados_requirements.m

@@ -18,7 +18,7 @@ function check_acados_requirements()
             error('Please set up CasADi yourself and try again.');
         else
             % download CasADi
-            CasADi_version = '3.4.5';
+            CasADi_version = '3.5.1';
             url = strcat('https://github.com/casadi/casadi/releases/download/',...
                    CasADi_version, '/');
             external_folder = fullfile(acados_dir, 'external');
@@ -35,7 +35,7 @@ function check_acados_requirements()
 
             if isunix
                 if verLessThan('matlab', '8.4')
-                    CasADi_version = '3.4.4';
+                    CasADi_version = '3.5.1';
                     filename = strcat('casadi-linux-matlabR2014a-v', CasADi_version, '.tar.gz');
                 else % R2014b or later
                     filename = strcat('casadi-linux-matlabR2014b-v', CasADi_version, '.tar.gz');

examples/acados_python/getting_started/mhe/minimal_example_mhe.py

@@ -131,4 +131,4 @@
 print('difference |x0_est - x0_bar|', np.linalg.norm(x0_bar - simXest[0, :]))
 print('difference |x_est - x_true|', np.linalg.norm(simXest - simX))
 
-plot_pendulum(h, Fmax, simU, simX, simXest, simY)
+plot_pendulum(h, Fmax, simU, simX, simXest, simY, latexify=False)

examples/acados_python/getting_started/ocp/minimal_example_ocp.py

@@ -95,10 +95,10 @@
 ocp.solver_options.hessian_approx = 'GAUSS_NEWTON'
 ocp.solver_options.integrator_type = 'ERK'
 ocp.solver_options.print_level = 0
+ocp.solver_options.nlp_solver_type = 'SQP' # SQP_RTI, SQP
 
 # set prediction horizon
 ocp.solver_options.tf = Tf
-ocp.solver_options.nlp_solver_type = 'SQP' # SQP_RTI
 
 ocp_solver = AcadosOcpSolver(ocp, json_file = 'acados_ocp.json')
 
@@ -116,4 +116,6 @@
     simU[i,:] = ocp_solver.get(i, "u")
 simX[N,:] = ocp_solver.get(N, "x")
 
-plot_pendulum(Tf/N, Fmax, simU, simX)
+ocp_solver.print_statistics() # encapsulates: stat = ocp_solver.get_stats("statistics")
+
+plot_pendulum(Tf/N, Fmax, simU, simX, latexify=False)

examples/acados_python/getting_started/ocp/ocp_example_cost_formulations.py

@@ -119,6 +119,7 @@
 
 ocp.solver_options.qp_solver = 'PARTIAL_CONDENSING_HPIPM' # FULL_CONDENSING_QPOASES
 ocp.solver_options.hessian_approx = HESSIAN_APPROXIMATION
+ocp.solver_options.regularize_method = 'CONVEXIFY'
 ocp.solver_options.integrator_type = 'ERK'
 
 ocp.solver_options.qp_solver_cond_N = 5
@@ -145,5 +146,5 @@
 simX[N,:] = ocp_solver.get(N, "x")
 
 
-plot_pendulum(Tf/N, Fmax, simU, simX)
+plot_pendulum(Tf/N, Fmax, simU, simX, latexify=False)
 

examples/acados_python/getting_started/sim/minimal_example_sim.py

@@ -83,4 +83,4 @@
     raise Exception('acados returned status {}. Exiting.'.format(status))
 
 # plot results
-plot_pendulum(Tf/N, 10, np.zeros((N, nu)), simX)
+plot_pendulum(Tf/N, 10, np.zeros((N, nu)), simX, latexify=False)