/home/laradmin/lar/perception/planarobstacles/mtt/src/mtt_association.cpp

Go to the documentation of this file.

/**************************************************************************************************
 Software License Agreement (BSD License)

 Copyright (c) 2011-2013, LAR toolkit developers - University of Aveiro - http://lars.mec.ua.pt
 All rights reserved.

 Redistribution and use in source and binary forms, with or without modification, are permitted
 provided that the following conditions are met:

  *Redistributions of source code must retain the above copyright notice, this list of
   conditions and the following disclaimer.
  *Redistributions in binary form must reproduce the above copyright notice, this list of
   conditions and the following disclaimer in the documentation and/or other materials provided
   with the distribution.
  *Neither the name of the University of Aveiro nor the names of its contributors may be used to
   endorse or promote products derived from this software without specific prior written permission.
 
 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR
 IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
 FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
 CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
 IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
 OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
***************************************************************************************************/
#include <mtt/mtt_association.h>

unsigned int last_id=0; 

void AssociateObjects(vector<t_listPtr> &list,vector<t_objectPtr> &objects,t_config& config,t_flag& flags)
{       
        for(uint i=0;i<objects.size();i++)
                objects[i]->object_found=false;
        
        double min_ret=1;
        int min_index=-1;
        double ret=1;
        bool association_found;
        double remove_threshold;
        
        for(uint i=0;i<list.size();i++)
        {
                min_ret=1;
                association_found=false;
                
                for(uint h=0;h<objects.size();h++)
                {
                        ret = CheckAssociationCriteria(*list[i],*objects[h]);
                        
                        if(ret<min_ret && ret<0) 
                        {
                                min_ret=ret;
                                min_index=h;
                                association_found=true;
                        }
                }
                
                //PFLN
                if(association_found)
                {
                        //                      double lret;
                        //                      double min_lret=1;
                        double dist;
                        double min_dist=1e6;
                        
                        int index_e=-1;
                        for(uint e=0;e<list.size();e++)
                        {
                                if(i==e)
                                        continue;
                                
                                dist=point2point_distance(objects[min_index]->cx,objects[min_index]->cy,list[e]->position.predicted_x,list[e]->position.predicted_y);
                                
                                if(dist<min_dist)
                                {
                                        min_dist=dist;
                                        index_e=e;
                                }
                        }
                        
                        //Exclusion zone B
                        //                      printf("%d %2.2f %2.2f\n",mlist->id,min_lret,min_ret);
                        
                        if(min_dist < config.ezB && list[index_e]->timers.lifetime > list[i]->timers.lifetime)
                        {
                                //                              printf("C%d A%d C%2.2f A%2.2f\n",mlist->id,aux2->id,min_ret,min_lret);
                                association_found=false;
                        }
                }
                
                //PFLN
                if(association_found && objects[min_index]->object_found==false)
                {
                        
                        if(list[i]->classification.partialy_occluded==false && objects[min_index]->partialy_occluded)
                        {
                                
                                objects[min_index]->cx=(objects[min_index]->cx+list[i]->position.predicted_x)/2;
                                objects[min_index]->cy=(objects[min_index]->cy+list[i]->position.predicted_y)/2;
                                
                        }
                        //PFLN
                        SingleObjectAssociation(*list[i],*objects[min_index]);
                        objects[min_index]->object_found=true;
                        list[i]->classification.occluded=false;
                        list[i]->classification.partialy_occluded=objects[min_index]->partialy_occluded;
                        
                        //PFLN
                }else
                {
                        //PFLN
                        
                        list[i]->classification.occluded=true;
                        list[i]->timers.occludedtime++;
                        
                        
                        list[i]->measurements.x=list[i]->position.predicted_x;
                        list[i]->measurements.y=list[i]->position.predicted_y;
                        
                        remove_threshold=list[i]->timers.lifetime;
                        
                        if(remove_threshold>config.max_missing_iterations)
                                remove_threshold=config.max_missing_iterations;
                        
                        if(list[i]->timers.occludedtime>remove_threshold)
                                RemoveFromList(list,list[i]->id);
                }
        }
                
        double dist_to_object=1e6;
        for(uint g=0;g<objects.size();g++)
        {
                objects[g]->min_distance_to_existing_object=1e6;
                
                for(uint i=0;i<list.size();i++)
                {
                        dist_to_object=point2point_distance(objects[g]->cx,objects[g]->cy,list[i]->measurements.x,list[i]->measurements.y);
                        
                        if(dist_to_object<objects[g]->min_distance_to_existing_object)
                                objects[g]->min_distance_to_existing_object=dist_to_object;
                }
                
                //Exclusion zone A
                if(objects[g]->min_distance_to_existing_object < config.ezA && flags.fi==false)
                        continue;
                
                if(objects[g]->object_found==false)
                        AddObjectToList(list,*objects[g],config);
        }
        
        return;
}

void RemoveFromList(vector<t_listPtr> &list,unsigned int id)
{
        vector<t_listPtr>::iterator it;
        
        for(it=list.begin();it!=list.end();it++)
        {
                if((*it)->id==id)
                {
                        cvReleaseKalman(&((*it)->motion_models.cv_x_kalman));
                        cvReleaseKalman(&((*it)->motion_models.cv_y_kalman));
                        cvReleaseKalman(&((*it)->motion_models.ca_x_kalman));
                        cvReleaseKalman(&((*it)->motion_models.ca_y_kalman));   

                        free((*it)->errors_cv.x_innovation);
                        free((*it)->errors_cv.x_residue);
                        free((*it)->errors_cv.y_innovation);
                        free((*it)->errors_cv.y_residue);
                        free((*it)->errors_cv.lateral_error);
                        //PFLN
                        free((*it)->errors_ca.x_innovation);
                        free((*it)->errors_ca.x_residue);
                        free((*it)->errors_ca.y_innovation);
                        free((*it)->errors_ca.y_residue);
                        free((*it)->errors_ca.lateral_error);

                        //PFLN
                        free((*it)->path_cv.x);
                        free((*it)->path_cv.y);
                        //PFLN
                        free((*it)->path_ca.x);
                        free((*it)->path_ca.y);
                        
                        list.erase(it);
                        return;
                }       
        }
}

void AllocErrors(t_errors*error,t_config&config)
{
        error->x_innovation=(double*)malloc(config.estimation_window_size*sizeof(double));
        
        error->x_residue=(double*)malloc(config.estimation_window_size*sizeof(double));
        
        error->y_innovation=(double*)malloc(config.estimation_window_size*sizeof(double));
        
        error->y_residue=(double*)malloc(config.estimation_window_size*sizeof(double));
        
        error->lateral_error=(double*)malloc(config.estimation_window_size*sizeof(double));
        
        error->x_inno_cov=0;
        error->x_resi_cov=0;
        error->y_inno_cov=0;
        error->y_resi_cov=0;
        error->lateral_error_cov=0;
        
        error->max_number_points=config.estimation_window_size;
        error->number_points=0;
        error->position=0;
        error->latest=0;
        error->next=1;
}

void SingleObjectAssociation(t_list& list,t_object& object)
{
        list.measurements.x=object.cx;
        list.measurements.y=object.cy;
        list.shape=object;
        
        SetOjectMorphology(list,object);
        object.object_found=true;
        object.id=list.id;
        
        list.timers.lifetime++;
        
        list.timers.occludedtime-=1;
        if(list.timers.occludedtime<0)
                list.timers.occludedtime=0;
}

void SetOjectMorphology(t_list&list,t_object& object)
{
        list.object_morphology.apparent_size=object.size;
}

void SetSearchArea(t_list& list,t_config&config)
{
        list.search_area.angle=list.velocity.velocity_angle;
        list.search_area.center_x=list.position.predicted_x;
        list.search_area.center_y=list.position.predicted_y;
        
        //      printf("%d %p Xp %2.2f Yp %2.2f\n",list->id,list,list->search_area.center_x,list->search_area.center_y);
        
        //      printf("%d EA %2.2f EB %2.2f\n",list->id,list->search_area.ellipse_A,list->search_area.ellipse_B);
        
        double size=list.object_morphology.apparent_size;
        double size_factor=0.1;
        
        double default_size=config.min_ellipse_axis;
        
        double xIl,yIl;
        
        if(list.model==CV)
        {
                xIl=list.errors_cv.x_inno_cov;
                yIl=list.errors_cv.y_inno_cov;
        }else
        {
                xIl=list.errors_ca.x_inno_cov;
                yIl=list.errors_ca.y_inno_cov;
        }
        
        int not_found_counter=list.timers.occludedtime;
        double not_found_factor=2;
        
        double innovation_error=sqrt(sqrt(xIl*xIl+yIl*yIl));
        double lateral_error=sqrt(list.errors_cv.lateral_error_cov);
        double A_innovation_factor,B_innovation_factor;
        double lateral_factor;
        
        if( list.velocity.velocity_module < 0.200 )
        {
                A_innovation_factor=5;
                B_innovation_factor=4;
                lateral_factor=1;
                
        }else
        {
                A_innovation_factor=5;
                B_innovation_factor=0.5;
                lateral_factor=4;
        }
        
        if(innovation_error<0.001)
                innovation_error=0.001;
        
        list.search_area.ellipse_A = not_found_factor*pow(not_found_counter,2) + A_innovation_factor*innovation_error + default_size + size_factor*size;
        list.search_area.ellipse_B = not_found_factor*pow(not_found_counter,2) + B_innovation_factor*innovation_error + default_size + size_factor*size + lateral_error*lateral_factor;
        
        if(list.search_area.ellipse_A > config.max_ellipse_axis)
                list.search_area.ellipse_A = config.max_ellipse_axis + default_size + size_factor*size;
        
        if(list.search_area.ellipse_B > config.max_ellipse_axis)
                list.search_area.ellipse_B = config.max_ellipse_axis + default_size + size_factor*size;
        
        if(list.timers.lifetime < 5)
        {
//              list.search_area.ellipse_A+=1.0;
//              list.search_area.ellipse_B+=1.0;
        }
}

void AddPointPath(t_path*path,double x,double y)
{
        path->x[path->position]=x;
        path->y[path->position]=y;
        
        path->latest=path->position;
        path->position++;
        
        if(path->position==path->max_number_points)
                path->position=0;
        
        path->next=path->position;
        
        if(path->number_points<path->max_number_points)
                path->number_points++;
}

double CheckAssociationCriteria(t_list&list,t_object& object)
{
        double ox=object.cx;
        double oy=object.cy;
        
        double angle=-list.search_area.angle;
        double s=list.search_area.ellipse_B;
        double r=list.search_area.ellipse_A;
        double M=cos(angle);
        double N=sin(angle);
        double c=list.search_area.center_x;
        double d=list.search_area.center_y;
        double tx=ox-c;
        double ty=oy-d;
        double A=(M*tx-N*ty)*(M*tx-N*ty);
        double B=(N*tx+M*ty)*(N*tx+M*ty);
        double Z=s*s*A+r*r*B-r*r*s*s;
        
        if(Z<0)
                return Z;
        else
                return 1;
}

void AddObjectToList(vector<t_listPtr> &list,t_object& object,t_config&config)
{
        t_listPtr element(new t_list);
                
        AllocMotionModels(*element,config);
        
        AllocPath(&(element->path_cv),config);
        AllocPath(&(element->path_ca),config);
        
        AllocErrors(&(element->errors_cv),config);
        AllocErrors(&(element->errors_ca),config);
        
        element->measurements.x=object.cx;
        element->measurements.y=object.cy;
        element->position.estimated_x=object.cx;
        element->position.estimated_y=object.cy;
        
        InitialiseSearchArea(*element,config);
        InitialiseTimers(&(element->timers));
        InitialiseClassification(&(element->classification));
        
        element->object_morphology.apparent_size=object.size;
        
        element->model=CV;
        element->shape=object;
        
        element->id=last_id;
//      element->branches=NULL;
//      element->next=NULL;
        
//      cout<<"id "<<element->id<<endl;
        last_id++;
        
        list.push_back(element);
        
        return;
}

void InitialiseClassification(t_classification*classification)
{
        classification->velocity_classification=MOVING;
        classification->occluded=false;
        classification->partialy_occluded=false;
}

void InitialiseTimers(t_timers*timer)
{
        timer->occludedtime=0;
        timer->lifetime=0;
}

void InitialiseSearchArea(t_list&list,t_config&config)
{
        list.search_area.ellipse_A=config.max_ellipse_axis;
        list.search_area.ellipse_B=config.max_ellipse_axis;
        list.search_area.angle=0;
        list.search_area.center_x=list.position.estimated_x;
        list.search_area.center_y=list.position.estimated_y;
}

void AllocPath(t_path*path,t_config&config)
{
        path->x=(double*)malloc(config.path_lenght*sizeof(double));
        
        path->y=(double*)malloc(config.path_lenght*sizeof(double));
        
        path->max_number_points=config.path_lenght;
        path->number_points=0;
        path->position=0;
        path->latest=0;
        path->next=1;
        
        return;
}

---

mtt
Author(s): Jorge Almeida
autogenerated on Wed Jul 23 04:34:57 2014