d-desiatkin
commented
4 years ago UPDATE
I recompiled openvslam with address sanitizers. Here is the output:
robotics@jetson-tx2:~/porter_ws/src/openvslam/build$ sudo ./run_euroc_slam -v ../ros/config/orb_vocab.dbow2 -d ~/Downloads/mav0/ -c ../example/euroc/EuRoC_mono.yaml --map-db /home/robotics/map.msg --debug [2020-05-08 19:27:59.388] [D] CONSTRUCT: config [2020-05-08 19:27:59.388] [I] config file loaded: ../example/euroc/EuRoC_mono.yaml [2020-05-08 19:27:59.389] [D] load camera model type [2020-05-08 19:27:59.389] [D] load camera model parameters [2020-05-08 19:27:59.389] [D] CONSTRUCT: camera::base [2020-05-08 19:27:59.389] [D] CONSTRUCT: camera::perspective [2020-05-08 19:27:59.389] [D] compute image bounds [2020-05-08 19:27:59.389] [D] load ORB parameters [2020-05-08 19:27:59.390] [D] load tracking parameters [2020-05-08 19:27:59.390] [D] load depth threshold [2020-05-08 19:27:59.390] [D] load depthmap factor [2020-05-08 19:27:59.475] [D] CONSTRUCT: system
/ \ _ \ \ / / | | /\ | \/ | | () | ' \/ -) ' \ V /__ \ |_ / | |\/| | _/| ./_||||_/ |_/__// __| || |_|
Copyright (C) 2019, National Institute of Advanced Industrial Science and Technology (AIST) All rights reserved.
This is free software, and you are welcome to redistribute it under certain conditions. See the LICENSE file.
Camera Configuration:
- name: EuRoC monocular
- setup: Monocular
- fps: 20
- cols: 752
- rows: 480
- color: Gray
- model: Perspective
- fx: 458.654
- fy: 457.296
- cx: 367.215
- cy: 248.375
- k1: -0.283408
- k2: 0.0739591
- p1: 0.00019359
- p2: 1.76187e-05
- k3: 0
- min x: -135.812
- max x: 895.507
- min y: -92.8751
- max y: 565.556 ORB Configuration:
- number of keypoints: 1000
- scale factor: 1.2
- number of levels: 8
- initial fast threshold: 20
minimum fast threshold: 7
[2020-05-08 19:27:59.476] [I] loading ORB vocabulary: ../ros/config/orb_vocab.dbow2 [2020-05-08 19:28:02.503] [D] CONSTRUCT: data::camera_database [2020-05-08 19:28:02.503] [D] CONSTRUCT: data::map_database [2020-05-08 19:28:02.503] [D] CONSTRUCT: data::bow_database [2020-05-08 19:28:02.505] [D] CONSTRUCT: publish::frame_publisher [2020-05-08 19:28:02.505] [D] CONSTRUCT: publish::map_publisher [2020-05-08 19:28:02.505] [D] CONSTRUCT: module::initializer [2020-05-08 19:28:02.505] [D] CONSTRUCT: module::relocalizer [2020-05-08 19:28:02.505] [D] CONSTRUCT: tracking_module [2020-05-08 19:28:02.505] [D] CONSTRUCT: mapping_module [2020-05-08 19:28:02.505] [D] CONSTRUCT: global_optimization_module [2020-05-08 19:28:02.505] [I] startup SLAM system [2020-05-08 19:28:02.508] [I] start mapping module [2020-05-08 19:28:02.508] [I] start global optimization module error: XDG_RUNTIME_DIR not set in the environment. [2020-05-08 19:28:23.582] [D] CONSTRUCT: initialize::perspective [2020-05-08 19:28:52.397] [I] initialization succeeded with F [2020-05-08 19:28:52.398] [D] DESTRUCT: initialize::perspective
==19059==ERROR: AddressSanitizer: attempting free on address which was not malloc()-ed: 0x007f2bfd27b0 in thread T8
0 0x7f9945777f in free (/usr/lib/aarch64-linux-gnu/libasan.so.4+0xcf77f)
#1 0x7f9716fe97 in g2o::JacobianWorkspace::~JacobianWorkspace() (/usr/local/lib/libg2o_core.so+0x49e97) #2 0x7f98ced91f in g2o::BlockSolver<g2o::BlockSolverTraits<6, 3> >::buildSystem() (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x41191f) #3 0x7f9716c2cf in g2o::OptimizationAlgorithmLevenberg::solve(int, bool) (/usr/local/lib/libg2o_core.so+0x462cf) #4 0x7f97164d5f in g2o::SparseOptimizer::optimize(int, bool) (/usr/local/lib/libg2o_core.so+0x3ed5f) #5 0x7f98de3483 in openvslam::optimize::global_bundle_adjuster::optimize(unsigned int, bool*) const (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x507483) #6 0x7f98c953e7 in openvslam::module::initializer::create_map_for_monocular(openvslam::data::frame&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x3b93e7) #7 0x7f98ca028b in openvslam::module::initializer::initialize(openvslam::data::frame&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x3c428b) #8 0x7f98a5a7af in openvslam::tracking_module::initialize() (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x17e7af) #9 0x7f98a65cb3 in openvslam::tracking_module::track() (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x189cb3) #10 0x7f98a67697 in openvslam::tracking_module::track_monocular_image(cv::Mat const&, double, cv::Mat const&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x18b697) #11 0x7f98a360ef in openvslam::system::feed_monocular_frame(cv::Mat const&, double, cv::Mat const&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x15a0ef) #12 0x55562fcb43 in mono_tracking(std::shared_ptr<openvslam::config> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, bool, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)::{lambda()#1}::operator()() const (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x27b43) #13 0x55562fd817 in std::thread::_State_impl<std::thread::_Invoker<std::tuple<mono_tracking(std::shared_ptr<openvslam::config> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, bool, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)::{lambda()#1}> > >::_M_run() (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x28817) #14 0x7f97c2550b (/usr/lib/aarch64-linux-gnu/libstdc++.so.6+0xcc50b) #15 0x7f988b7087 in start_thread (/lib/aarch64-linux-gnu/libpthread.so.0+0x7087)0x007f2bfd27b0 is located 16 bytes inside of 112-byte region [0x007f2bfd27a0,0x007f2bfd2810) allocated by thread T8 here:
0 0x7f99457b1b in malloc (/usr/lib/aarch64-linux-gnu/libasan.so.4+0xcfb1b)
#1 0x7f98ced1ff in Eigen::internal::aligned_malloc(unsigned long) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x4111ff) #2 0x7f98ced5d3 in g2o::BlockSolver<g2o::BlockSolverTraits<6, 3> >::buildSystem() (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x4115d3) #3 0x7f9716c2cf in g2o::OptimizationAlgorithmLevenberg::solve(int, bool) (/usr/local/lib/libg2o_core.so+0x462cf) #4 0x7f97164d5f in g2o::SparseOptimizer::optimize(int, bool) (/usr/local/lib/libg2o_core.so+0x3ed5f) #5 0x7f98de3483 in openvslam::optimize::global_bundle_adjuster::optimize(unsigned int, bool*) const (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x507483) #6 0x7f98c953e7 in openvslam::module::initializer::create_map_for_monocular(openvslam::data::frame&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x3b93e7) #7 0x7f98ca028b in openvslam::module::initializer::initialize(openvslam::data::frame&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x3c428b) #8 0x7f98a5a7af in openvslam::tracking_module::initialize() (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x17e7af) #9 0x7f98a65cb3 in openvslam::tracking_module::track() (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x189cb3) #10 0x7f98a67697 in openvslam::tracking_module::track_monocular_image(cv::Mat const&, double, cv::Mat const&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x18b697) #11 0x7f98a360ef in openvslam::system::feed_monocular_frame(cv::Mat const&, double, cv::Mat const&) (/home/robotics/porter_ws/src/openvslam/build/lib/libopenvslam.so+0x15a0ef) #12 0x55562fcb43 in mono_tracking(std::shared_ptr<openvslam::config> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, bool, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)::{lambda()#1}::operator()() const (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x27b43) #13 0x55562fd817 in std::thread::_State_impl<std::thread::_Invoker<std::tuple<mono_tracking(std::shared_ptr<openvslam::config> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, bool, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)::{lambda()#1}> > >::_M_run() (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x28817) #14 0x7f97c2550b (/usr/lib/aarch64-linux-gnu/libstdc++.so.6+0xcc50b) #15 0x7f988b7087 in start_thread (/lib/aarch64-linux-gnu/libpthread.so.0+0x7087) #16 0x7f97aab4eb (/lib/aarch64-linux-gnu/libc.so.6+0xd04eb)Thread T8 created by T0 here:
0 0x7f993c09a7 in pthread_create (/usr/lib/aarch64-linux-gnu/libasan.so.4+0x389a7)
#1 0x7f97c25813 in std::thread::_M_start_thread(std::unique_ptr<std::thread::_State, std::default_delete<std::thread::_State> >, void (*)()) (/usr/lib/aarch64-linux-gnu/libstdc++.so.6+0xcc813) #2 0x55562fe3af in mono_tracking(std::shared_ptr<openvslam::config> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, bool, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool) (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x293af) #3 0x55562f6dbb in main (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x21dbb) #4 0x7f979fb6df in __libc_start_main (/lib/aarch64-linux-gnu/libc.so.6+0x206df) #5 0x55562f7bff (/home/robotics/porter_ws/src/openvslam/build/run_euroc_slam+0x22bff)SUMMARY: AddressSanitizer: bad-free (/usr/lib/aarch64-linux-gnu/libasan.so.4+0xcf77f) in free ==19059==ABORTING
sudo-panda
System setup:
Jetson TX2 - Jetpack v 4.3 - Ubuntu 18.04 OpenCV v 3.4.9 g2o - I use commit specified in installation guide DBoW2 - I use your custom version Pangolin viewer - also from installation guide.
Without sudo:
And here all just freeze.
With sudo:
Here everything is fine.
Basically I have the problem with launching ros node. I found this behavior and can't make it work. I mean I successfully attach camera and see image in pangolin viewer, but when I shift my camera and first point appears all just freeze. It seems like node don't have enough rights to create a map. Do somebody have any idea why it could happen?
g2o was compiled with and without BUILD_WITH_MARCH_NATIVE