Questions tagged [slam]

SLAM (Simultaneous Localization And Mapping) refers to a robot building a map of its environment through it's sensor data (mapping) and keeping track of its own position in that map (localization) at the same time.

Filter by
Sorted by
Tagged with
0 votes
1 answer
40 views

What should I use for a Visual+IMU+GPS fusion? A Madgwick or a Kalman?

I am new to robotics, and recently I am involving in a sensor fusion task using visual input (binocular at present), an IMU, and a GPS module. I have searched for related journal papers for a ...
0 votes
0 answers
22 views

LSD-SLAM vs Stereo Vision

LSD-SLAM is a monoecular SLAM system which is considered "direct", as in it does not find features explicitly like in ORB slam. The paper is available here: https://link.springer.com/chapter/...
1 vote
0 answers
30 views

How does marginalization work in fixed lag smoothing?

I have read in SLAM papers talk about marginaliztion. I am confused about how this process works, why you do it and how it differs from global smoothing. In particular if you're interested in Windowed ...
1 vote
1 answer
50 views

How to use hector slam autonomous navigation without odometry? [closed]

I have a robot that i want to make autonomous, i will try to get a rplidar A1 for making things easy on the mapping through the HECTOR SLAM package but i want to use autonomous navigation, i only have ...
0 votes
0 answers
12 views

Mono SFM and moving objects

When building a map is there a reliable way to deal with moving objects? Particularly for mono inertial SLAM/structure from motion?
4 votes
2 answers
114 views

Why do we need to know the translation between Camera and IMU?

In Visual Inertial Odometry, assuming the camera and the IMU are attached to the same rigid body, why isn't it enough to know just the relative rotation between the camera and the IMU? Why do we need ...
  • 63
0 votes
0 answers
15 views

Bundle Adjustment and error propagation

When you do bundle adjustment do states with higher accuracy have stronger certainty during the optimization? For example a state which was estimates right after starting will be closer to the true ...
0 votes
1 answer
24 views

SLAM: global BA and IMU rate

My understanding is that in SLAM systems keyframes are recorded at a lower rate and used for loop closing and global BA vs every every image which is used for a sliding window optimization. What I am ...
0 votes
0 answers
23 views

Mono Slam scale tracking

How is the scale tracked in monoecular slam algorithms between image pairs? It appears the 5point algorithm is used to initially compute the first two cameras poses but after that you want to maintain ...
0 votes
0 answers
34 views

Stereo Bundle Adjustment

When doing stereo based visual odometry and bundle adjustment how does the stereo constraint get factored into the photometric error? Do you use the fact the second cameras location is known wrt the ...
0 votes
1 answer
35 views

Question about bundle adjustment

When doing bundle adjustment do landmarks gets optimized as stand alone variables or are they viewed as a function of the robots first position when they are detected?
0 votes
1 answer
123 views

Transform problem with rviz map display when using the SLAMTEC Mapper M2M2 with ROS

I am working on a project using the SLAMTEC Mapper M2M2 with ROS to map an area of interest. I am using Oracle Virtualbox to run ROS Noetic with Ubuntu 20.04, and my host machine is running on Windows ...
0 votes
0 answers
18 views

Synchronize GPS orientation with cameras

I'm interested in using GPS system to generate ground truth for evaluation of some SLAM systems I am testing. I'm confused on how I can initialize the location and orientation of the GPS unit to be ...
0 votes
1 answer
49 views

Why would you use direct SLAM over feature-based?

I am curious, why would you ever choose direct slam(photometric consistency, etc) over feature based slam? Particularly for dynamic scenes? It seems you're always better off aligning fewer points ...
0 votes
1 answer
38 views

How does dense mapping/voxel occupancy grid map work with bundle adjustment?

When running for example some version of Orb Slam with stereo vision you're using sparse features for tracking but you can also infer a full depth map. I am curious how you use that full depth map to ...
1 vote
2 answers
75 views

SLAM with multiple cameras and asynchronous sensors

I'm pretty sure this is the right place for this question, but if not, please point me in the right direction! Basically, I would like to know how do SLAM using multiple cameras and sensors that are ...
1 vote
2 answers
85 views

Estimating angular velocity from position

If you have two position estimates $P_{t_1}$ and $P_{t_2}$ and times $t_1$ and $t_2$ how do you extract the average angular velocity from them? The linear velocity is a simple difference but how do ...
0 votes
2 answers
34 views

How to read the map output file generated by a project using Cartographer without ROS

Currently I can successfully generate the map output file but I have no idea how to display it in practice. Basically, it should be a 2D SLAM map. The project I'm using is available at the following ...
  • 103
1 vote
2 answers
134 views

google cartographer

can we use google cartographer without installing ros, like only using c++ or python? I have tried looking for c++ but did not get how to use, but in some website it is shown we can see it without ros
0 votes
0 answers
106 views

IMU to Camera transformation

Suppose you are implementing SLAM or VIO system, you have your camera C and IMU and a coordinate transformation between them (R, T). How do convert IMU measurements to camera accelerations and ...
1 vote
1 answer
101 views

Why have EKF and PF slam algorithms been abandoned?

As far as I know all current 3d SLAM systems use pose graph optimization, both local and global. EKF and particle filter algorithms are not used or referenced at all as far as I can tell and the ...
0 votes
0 answers
111 views

error with Breezyslam (Descriptor length mismatch)

so I was working on a slam project and using the Breezyslam repo link, I am always getting this error, I don't know where I am getting it wrong, can someone help? ...
0 votes
0 answers
99 views

OpenCV: What does it mean when the number of inliers returned by recoverPose() function is 0?

I've been working on a pose estimation project and one of the steps is finding the pose using the recoverPose function of OpenCV. ...
  • 205
0 votes
0 answers
52 views

EKF slam loop closure and landmark location updates

I asked a variation of the question before but I never got an answer. Basically I'm trying to understand how after loop closure the location of prior seen landmarks gets updated and their uncertainty ...
0 votes
0 answers
35 views

Question about 4-mecanum wheeled mobile robots

I'm trying to make a mobile robot with 4 mecanum wheels. The method that I've chosen is to fuse the sensors(encoder, imu) with EKF and use the SLAM algorithms(Gmapping, Cartographer, etc..). Right now ...
  • 1
1 vote
0 answers
487 views

How to use Nav2, slam_toolbox with odom data, gps and imu sensors?

Setup details: ROS2 foxy on amd64 architecture CPU with nav2 and slam_toolbox installed Robot is Clearpath husky with Velodyne VLP-16 lidar, IMU and GPS sensor in gazebo. Can see data on every topic ...
2 votes
1 answer
112 views

Pipeline for dense reconstruction using pose estimation from orb-slam and stereo camera

I am trying to perform dense reconstruction using a sequence of images from a calibrated stereo camera. I have been using orb-slam3 to give me the camera's pose estimation. I am also generating the ...
  • 23
0 votes
1 answer
51 views

Localization by comparing current lidar scan with previous lidar scan

I have managed to use an ICP algorithm to produce a relative pose difference between a new lidar range scan and the previous lidar scan. When I tested it on individual scan pairs, the results look ...
0 votes
0 answers
38 views

Which scan matching technique is most suitable for 2D laser range data?

I have odometry data [x, y, theta] and 2D laser range data from a robot moving in a building. Based on a new scan, I want to match it with a previous scan to get a corrected pose estimation for the ...
0 votes
1 answer
67 views

How to perform time-offset calibration use GTSAM in Camera-IMU system?

Assume we have camera poses data and IMU data(which include gyr and acc data) with different timestamps, now I want to estimate this value, I known that this can be done in many ways(EKF,MSCKF), but I ...
  • 1
0 votes
0 answers
19 views

Robotics processing pipeline that is agnostic of ROS or other data source

I'm rather new to the field of robotics and am wondering how to go about setting up my system architecture. I'm trying to build a SLAM pipeline (initially very simple based on 2D point clouds, with ...
0 votes
0 answers
41 views

Loop closure implementation in FastSLAM 1.0

I have implemented the FastSLAM 1.0 algorithm in C++, the algorithm takes in 2D position estimates of landmarks and odometry data from the combination of IMU, GPS and wheel speed odometry data. The ...
0 votes
0 answers
9 views

Why can we perform the following threshold check to see if we are current at zero velocity in Inertial-based Detection?

recently I was working on works using open-vins, which is very helpful for new-comers, but when I'm read docs, some question occurs to me, In this page zero velocity update, under zero velocity ...
  • 1
0 votes
0 answers
110 views

Adding/ Subtracting vertex and edge constraints using g2o dataset in a monocular odometry project

I have come across a relatively new rotation averaging solver with full python implementation and helpful examples https://github.com/dellaert/ShonanAveraging However, I am having a hard time ...
0 votes
0 answers
30 views

Can we use neural networks to solve the problem of SLAM? and how?

As a neural network can estimate almost any arbitrary function, I was wondering if we can use it to replace traditional SLAM algorithm. If that is possible, please explain how exactly you think it can ...
  • 101
0 votes
0 answers
37 views

EKF slam Landmark covariance propagation

Is there an intuitive explanation for how loop closure reduces the landmarks of prior observations? In particular the landmarks are connected to each other only through the observer, observing a ...
0 votes
1 answer
47 views

How to deal with inexact pointcloud matching?

When running ICP(for example with SLAM) it's possible your next position does not have a point cloud which matches up with the prior point cloud due to resolution limits. How is this dealt with it? ...
0 votes
1 answer
53 views

Derivation of EKF slam landmark initialization

I have read a couple posts which gives the formula for initializing a new landmark in EKF slam but the derivation is not given. EKF-SLAM initialize new landmark in covariance matrix Could someone ...
0 votes
1 answer
112 views

Incorporating motion model in Graph Slam and Bundle adjustment

When doing global bundle adjustment how does one incorporate a motion model or IMU information? Does it simply impose a cost function in relation to how far away the expected location is of the robot ...
0 votes
1 answer
138 views

EKF slam vs global bundle adjustment

How does EKF slam and the covariance matrix interact with global bundle adjustment? In particular the covariance matrix gives you a confidence on location but global bundle adjustment does not. How do ...
1 vote
2 answers
469 views

/rtabmap/rtabmap: Did not receive data since 5 seconds

trying to run RtabMap on my realsense-435 device. I've set up all the topics and made sure they are published and subscribed, nevertheless, RtabMap gives me this error: ...
  • 51
0 votes
1 answer
45 views

How to do path planning using only a fixed number of points?

I have a vehicle with cameras to detect different colored cones. I want to generate a drivable path based on the location of the detected cones.The number of detected cones varies, as it depends on ...
2 votes
0 answers
54 views

Monocular SLAM position estimation

I have been reading about mono SLAM and I have a couple questions which are not so clear: When doing pose estimation, I have seen people both reference homography transformation and also reference 5\...
1 vote
1 answer
376 views

slam_toolbox mapping is not life-long?

I've tested slam_toolbox producing life-long environment mapping, and not quite satisfied with the results. Below you can see a fragment of the mapping. I used a 1x0.5m case to test the changing map ...
  • 15
0 votes
1 answer
64 views

How to make gmapping dynamic, or advise any other methods to create mapping of a dynamic environment

it is a general question. Gmapping seems to be used only for creating a map of the static environment. Once created, it doesn't account for any changes - you add a new object, but it's being ignored ...
  • 15
0 votes
0 answers
44 views

How is loop closure mechanism incoporrated in particle filter based SLAM algorithm?

Could any one please explain, how is loop clousre mechanism introduced in filtering based SLAM techniques like FAST Slam ?
1 vote
1 answer
62 views

How to get from lidar coordinates to gps coordinates

I build a map from lidar slam lego-loam, which is a lidar odometer, and then I want to use this map by gps localization. I know the map will have errors in some ...
  • 111
0 votes
0 answers
40 views

How to calculate the covariance and gain in SLAM when only one measurement is available?

I am trying to perform SLAM for cases where only one sensor measurement is available. For example, suppose I want to track the position of a robot moving in a room with multiple known landmarks (2D ...
0 votes
1 answer
58 views

Find a mistake in transformation between different coordinate frames in the WIP-calibration pipeline

I design the extrinsics calibration pipeline for the robotic agents, carrying onboard two sensors of different modalities. I aim to calibrate one of them (S) ...
  • 131
0 votes
0 answers
35 views

offset in gridmap

colleagues, i am currently viewing Cyrill Stachniss's course on SLAM and also doing the exercises. At the moment i am doing an exercise on occupancy grid mapping and there is some strange offset ...
  • 1

1
2 3 4 5
8