開貼記錄一下學習與應用scan context的經驗。（由於苦於尋找segmatch的odom，然后用loam做odom重定位效果並不是很理想，遂轉戰sacn context。）
一切不以應用為目的經驗貼都是耍流氓
1.應用
其應用比較簡單，只需在函數中加入
SCManager scManager;//初始化
auto 結果 = scManager.detectLoopClosureID();//返回兩個值
scManager.makeAndSaveScancontextAndKeys(*點雲);//其中點雲為pcl::PointCloud<pcl::PointXYZI> & cloud_name
對於其兩個返回值有
SCclosestHistoryFrameID = 結果.first;//第一個值是幀ID
yawDiffRad = 結果.second; //第二個值是航向偏差 由序號得出將360度分為60份
2.源碼解析
float rad2deg(float radians)
float deg2rad(float degrees)
float xy2theta( const float & _x, const float & _y )// xy2theta
MatrixXd circshift( MatrixXd &_mat, int _num_shift )//將mat陣行右移得到新矩陣num shift
{
// shift columns to right direction 將列右移
assert(_num_shift >= 0);//如果表達式為FALSE (0), 程序將報告錯誤，並終止執行。如果表達式不為0，則繼續執行后面的語句
if( _num_shift == 0 ) {
MatrixXd shifted_mat( _mat );//預定義
return shifted_mat; // Early return
}
MatrixXd shifted_mat = MatrixXd::Zero( _mat.rows(), _mat.cols() );//初始化
for ( int col_idx = 0; col_idx < _mat.cols(); col_idx++ ){
int new_location = (col_idx + _num_shift) % _mat.cols();//行右移輸入量num shift
shifted_mat.col(new_location) = _mat.col(col_idx);//shifted新矩陣賦值
}
return shifted_mat;
} // circshift
double SCManager::distDirectSC ( MatrixXd &_sc1, MatrixXd &_sc2 )
{
int num_eff_cols = 0; // i.e., to exclude all-nonzero sector
double sum_sector_similarity = 0;
for ( int col_idx = 0; col_idx < _sc1.cols(); col_idx++ ){
VectorXd col_sc1 = _sc1.col(col_idx);//預定義
VectorXd col_sc2 = _sc2.col(col_idx);
if( col_sc1.norm() == 0 | col_sc2.norm() == 0 )
continue; // don't count this sector pair. 0，0的時候不計算
double sector_similarity = col_sc1.dot(col_sc2) / (col_sc1.norm() * col_sc2.norm());//點積/叉積 = cos
sum_sector_similarity = sum_sector_similarity + sector_similarity;//計算總的
num_eff_cols = num_eff_cols + 1;
}
double sc_sim = sum_sector_similarity / num_eff_cols;
return 1.0 - sc_sim;
} // distDirectSC
int SCManager::fastAlignUsingVkey( MatrixXd & _vkey1, MatrixXd & _vkey2)
{
int argmin_vkey_shift = 0;
double min_veky_diff_norm = 10000000;
for ( int shift_idx = 0; shift_idx < _vkey1.cols(); shift_idx++ ){
MatrixXd vkey2_shifted = circshift(_vkey2, shift_idx);//行變換
MatrixXd vkey_diff = _vkey1 - vkey2_shifted;//減去shifted
double cur_diff_norm = vkey_diff.norm();//歐幾里得長度 或 模
if( cur_diff_norm < min_veky_diff_norm ){
argmin_vkey_shift = shift_idx; //第幾個
min_veky_diff_norm = cur_diff_norm; //更新最小值
}
}
return argmin_vkey_shift;//返回第幾個
} // fastAlignUsingVkey
std::pair<double, int> SCManager::distanceBtnScanContext( MatrixXd &_sc1, MatrixXd &_sc2 )
{
// 1. fast align using variant key (not in original IROS18)
MatrixXd vkey_sc1 = makeSectorkeyFromScancontext( _sc1 );
MatrixXd vkey_sc2 = makeSectorkeyFromScancontext( _sc2 );//返回一個值
int argmin_vkey_shift = fastAlignUsingVkey( vkey_sc1, vkey_sc2 );//返回第幾個 由歐式距離
const int SEARCH_RADIUS = round( 0.5 * SEARCH_RATIO * _sc1.cols() ); // a half of search range //設置搜索半徑
std::vector<int> shift_idx_search_space { argmin_vkey_shift };
for ( int ii = 1; ii < SEARCH_RADIUS + 1; ii++ ){
shift_idx_search_space.push_back( (argmin_vkey_shift + ii + _sc1.cols()) % _sc1.cols() ); //把范圍內的加進去
shift_idx_search_space.push_back( (argmin_vkey_shift - ii + _sc1.cols()) % _sc1.cols() );
}
std::sort(shift_idx_search_space.begin(), shift_idx_search_space.end()); //排個序
// 2. fast columnwise diff
int argmin_shift = 0;
double min_sc_dist = 10000000;
for ( int num_shift: shift_idx_search_space ){
MatrixXd sc2_shifted = circshift(_sc2, num_shift); //行變換
double cur_sc_dist = distDirectSC( _sc1, sc2_shifted ); //計算向量之類的
if( cur_sc_dist < min_sc_dist ){
argmin_shift = num_shift;
min_sc_dist = cur_sc_dist;//更新
}
}
return make_pair(min_sc_dist, argmin_shift);
} // distanceBtnScanContext
MatrixXd SCManager::makeScancontext( pcl::PointCloud<SCPointType> & _scan_down )
{
TicToc t_making_desc;
int num_pts_scan_down = _scan_down.points.size();//輸入點的數量
// main
const int NO_POINT = -1000;
MatrixXd desc = NO_POINT * MatrixXd::Ones(PC_NUM_RING, PC_NUM_SECTOR);//預定義
SCPointType pt;
float azim_angle, azim_range; // wihtin 2d plane
int ring_idx, sctor_idx;
for (int pt_idx = 0; pt_idx < num_pts_scan_down; pt_idx++) {
pt.x = _scan_down.points[pt_idx].x;
pt.y = _scan_down.points[pt_idx].y;
pt.z = _scan_down.points[pt_idx].z + LIDAR_HEIGHT; // naive adding is ok (all points should be > 0).雷達自身高2.0米
// xyz to ring, sector
azim_range = sqrt(pt.x * pt.x + pt.y * pt.y);//半徑
azim_angle = xy2theta(pt.x, pt.y); //方向角
// if range is out of roi, pass
if( azim_range > PC_MAX_RADIUS )//設置一個半徑的濾波
continue;
ring_idx = std::max( std::min( PC_NUM_RING, int(ceil( (azim_range / PC_MAX_RADIUS) * PC_NUM_RING )) ), 1 );//線號
sctor_idx = std::max( std::min( PC_NUM_SECTOR, int(ceil( (azim_angle / 360.0) * PC_NUM_SECTOR )) ), 1 );//角度的扇號
// taking maximum z
if ( desc(ring_idx-1, sctor_idx-1) < pt.z ) // -1 means cpp starts from 0
desc(ring_idx-1, sctor_idx-1) = pt.z; // update for taking maximum value at that bin//更新Z最小值
}
// reset no points to zero (for cosine dist later)
for ( int row_idx = 0; row_idx < desc.rows(); row_idx++ )
for ( int col_idx = 0; col_idx < desc.cols(); col_idx++ )//行列
if( desc(row_idx, col_idx) == NO_POINT )
desc(row_idx, col_idx) = 0;//若沒有點則置0
t_making_desc.toc("PolarContext making");
return desc;
} // SCManager::makeScancontext
MatrixXd SCManager::makeRingkeyFromScancontext( Eigen::MatrixXd &_desc )
{
/*
* summary: rowwise mean vector
*/
Eigen::MatrixXd invariant_key(_desc.rows(), 1);//預定義
for ( int row_idx = 0; row_idx < _desc.rows(); row_idx++ )
{
Eigen::MatrixXd curr_row = _desc.row(row_idx);
invariant_key(row_idx, 0) = curr_row.mean();//均值 行 //16線
}
return invariant_key;
} // SCManager::makeRingkeyFromScancontext
MatrixXd SCManager::makeSectorkeyFromScancontext( Eigen::MatrixXd &_desc )
{
//summary: columnwise mean vector
Eigen::MatrixXd variant_key(1, _desc.cols()); //預定義
for ( int col_idx = 0; col_idx < _desc.cols(); col_idx++ )
{
Eigen::MatrixXd curr_col = _desc.col(col_idx);//行
variant_key(0, col_idx) = curr_col.mean(); //求均值 列 //360°
}
return variant_key;
} // SCManager::makeSectorkeyFromScancontext
void SCManager::makeAndSaveScancontextAndKeys( pcl::PointCloud<SCPointType> & _scan_down )
{
Eigen::MatrixXd sc = makeScancontext(_scan_down); // v1 行列設置
Eigen::MatrixXd ringkey = makeRingkeyFromScancontext( sc ); //返回第一個值
Eigen::MatrixXd sectorkey = makeSectorkeyFromScancontext( sc ); //返回另一個值
std::vector<float> polarcontext_invkey_vec = eig2stdvec( ringkey );//線號
polarcontexts_.push_back( sc );
polarcontext_invkeys_.push_back( ringkey );
polarcontext_vkeys_.push_back( sectorkey );
polarcontext_invkeys_mat_.push_back( polarcontext_invkey_vec );
// cout <<polarcontext_vkeys_.size() << endl;
} // SCManager::makeAndSaveScancontextAndKeys
std::pair<int, float> SCManager::detectLoopClosureID ( void )
{
int loop_id { -1 }; // init with -1, -1 means no loop (== LeGO-LOAM's variable "closestHistoryFrameID")
auto curr_key = polarcontext_invkeys_mat_.back(); // current observation (query)線號
auto curr_desc = polarcontexts_.back(); // current observation (query)//行列號
/*
* step 1: candidates from ringkey tree_
*/
if( polarcontext_invkeys_mat_.size() < NUM_EXCLUDE_RECENT + 1)// 線號size <51 從0到50
{
std::pair<int, float> result {loop_id, 0.0}; //構造result
return result; // Early return
}
// tree_ reconstruction (not mandatory to make everytime)
if( tree_making_period_conter % TREE_MAKING_PERIOD_ == 0) // to save computation cost // %50
{
TicToc t_tree_construction;
polarcontext_invkeys_to_search_.clear(); //
polarcontext_invkeys_to_search_.assign( polarcontext_invkeys_mat_.begin(), polarcontext_invkeys_mat_.end() - NUM_EXCLUDE_RECENT ) ;//用於拷貝、賦值操作
polarcontext_tree_.reset(); //智能指針 kdtree適應
polarcontext_tree_ = std::make_unique<InvKeyTree>(PC_NUM_RING /* dim */, polarcontext_invkeys_to_search_, 10 /* max leaf */ );//賦值
// tree_ptr_->index->buildIndex(); // inernally called in the constructor of InvKeyTree (for detail, refer the nanoflann and KDtreeVectorOfVectorsAdaptor)
t_tree_construction.toc("Tree construction");
}
tree_making_period_conter = tree_making_period_conter + 1;
double min_dist = 10000000; // init with somthing large
int nn_align = 0;
int nn_idx = 0;
// knn search
std::vector<size_t> candidate_indexes( NUM_CANDIDATES_FROM_TREE );
std::vector<float> out_dists_sqr( NUM_CANDIDATES_FROM_TREE );
TicToc t_tree_search;
nanoflann::KNNResultSet<float> knnsearch_result( NUM_CANDIDATES_FROM_TREE ); //預定義
knnsearch_result.init( &candidate_indexes[0], &out_dists_sqr[0] );//初始化
polarcontext_tree_->index->findNeighbors( knnsearch_result, &curr_key[0] /* query */, nanoflann::SearchParams(10) );
t_tree_search.toc("Tree search");
/*
* step 2: pairwise distance (find optimal columnwise best-fit using cosine distance)
*/
TicToc t_calc_dist;
for ( int candidate_iter_idx = 0; candidate_iter_idx < NUM_CANDIDATES_FROM_TREE; candidate_iter_idx++ )
{
MatrixXd polarcontext_candidate = polarcontexts_[ candidate_indexes[candidate_iter_idx] ];//行列號
std::pair<double, int> sc_dist_result = distanceBtnScanContext( curr_desc, polarcontext_candidate );//最小距離 和 序號
double candidate_dist = sc_dist_result.first;
int candidate_align = sc_dist_result.second;
if( candidate_dist < min_dist )
{
min_dist = candidate_dist;
nn_align = candidate_align;
nn_idx = candidate_indexes[candidate_iter_idx];// 更新
}
}
t_calc_dist.toc("Distance calc");
/*
* loop threshold check
*/
if( min_dist < SC_DIST_THRES )
{
loop_id = nn_idx;
// std::cout.precision(3);
cout << "[Loop found] Nearest distance: " << min_dist << " btn " << polarcontexts_.size()-1 << " and " << nn_idx << "." << endl;
cout << "[Loop found] yaw diff: " << nn_align * PC_UNIT_SECTORANGLE << " deg." << endl;
}
else
{
std::cout.precision(3);
cout << "[Not loop] Nearest distance: " << min_dist << " btn " << polarcontexts_.size()-1 << " and " << nn_idx << "." << endl;
cout << "[Not loop] yaw diff: " << nn_align * PC_UNIT_SECTORANGLE << " deg." << endl;
}
// To do: return also nn_align (i.e., yaw diff)
float yaw_diff_rad = deg2rad(nn_align * PC_UNIT_SECTORANGLE);
std::pair<int, float> result {loop_id, yaw_diff_rad};
return result;
} // SCManager::detectLoopClosureID
// } // namespace SC2
3.應用開發
pcd文件需要分割然后進行scancontext特征的抽取
對PCD文件數據進行解析，希望找到跳變沿
4037 1442 712 4227 4437 6097 8724 14062在這些地方發生了20以上的跳變或者增減順序的跳變（-20增加到50 突然變成-20或者變為20），並無任何順序
后面准備在建圖時將scancontext信息保存下來，用來做后續的全局定位