Visual odometry | EPFL Graph Search

In robotics and computer vision, visual odometry is the process of determining the position and orientation of a robot by analyzing the associated camera images. It has been used in a wide variety of robotic applications, such as on the Mars Exploration Rovers. Overview In navigation, odometry is the use of data from the movement of actuators to estimate change in position over time through devices such as rotary encoders to measure wheel rotations. While useful for many wheeled or tracked vehicles, traditional odometry techniques cannot be applied to mobile robots with non-standard locomotion methods, such as legged robots. In addition, odometry universally suffers from precision problems, since wheels tend to slip and slide on the floor creating a non-uniform distance traveled as compared to the wheel rotations. The error is compounded when the vehicle operates on non-smooth surfaces. Odometry readings become increasingly unreliable as these errors accumulate and compound

VIODE: A Simulated Dataset to Address the Challenges of Visual-Inertial Odometry in Dynamic Environments

Dario Floreano, Koji Minoda, Fabian Maximilian Schilling, Valentin Wüest

Dynamic environments such as urban areas are still challenging for popular visual-inertial odometry (VIO) algorithms. Existing datasets typically fail to capture the dynamic nature of these environments, therefore making it difficult to quantitatively evaluate the robustness of existing VIO methods. To address this issue, we propose three contributions: firstly, we provide the VIODE benchmark, a novel dataset recorded from a simulated UAV that navigates in challenging dynamic environments. The unique feature of the VIODE dataset is the systematic introduction of moving objects into the scenes. It includes three environments, each of which is available in four dynamic levels that progressively add moving objects. The dataset contains synchronized stereo images and IMU data, as well as ground-truth trajectories and instance segmentation masks. Secondly, we compare state-of-the-art VIO algorithms on the VIODE dataset and show that they display substantial performance degradation in highly dynamic scenes. Thirdly, we propose a simple extension for visual localization algorithms that relies on semantic information. Our results show that scene semantics are an effective way to mitigate the adverse effects of dynamic objects on VIO algorithms. Finally, we make the VIODE dataset publicly available at https://github.com/kminoda/VIODE.

2021

Robot Identification and Localization with Pointing Gestures

We propose a novel approach to establish the relative pose of a mobile robot with respect to an operator that wants to interact with it; we focus on scenarios in which the robot is in the same environment as the operator, and is visible to them. The approach is based on comparing the trajectory of the robot, which is known in the robot's odometry frame, to the motion of the arm of the operator, who, for a short time, keeps pointing at the robot they want to interact with. In multi-robot scenarios, the same approach can be used to simultaneously identify which robot the operator wants to interact with. The main advantage over alternatives is that our system only relies on the robot's odometry, on a wearable inertial measurement unit (IMU), and, crucially, on the operator's own perception. We experimentally show the feasibility of our approach using real-world robots.

2019

3D position tracking for all-terrain robots

Pierre Lamon

Rough terrain robotics is a fast evolving field of research and a lot of effort is deployed towards enabling a greater level of autonomy for outdoor vehicles. Such robots find their application in scientific exploration of hostile environments like deserts, volcanoes, in the Antarctic or on other planets. They are also of high interest for search and rescue operations after natural or artificial disasters. The challenges to bring autonomy to all terrain rovers are wide. In particular, it requires the development of systems capable of reliably navigate with only partial information of the environment, with limited perception and locomotion capabilities. Amongst all the required functionalities, locomotion and position tracking are among the most critical. Indeed, the robot is not able to fulfill its task if an inappropriate locomotion concept and control is used, and global path planning fails if the rover loses track of its position. This thesis addresses both aspects, a) efficient locomotion and b) position tracking in rough terrain. The Autonomous System Lab developed an off-road rover (Shrimp) showing excellent climbing capabilities and surpassing most of the existing similar designs. Such an exceptional climbing performance enables an extension in the range of possible areas a robot could explore. In order to further improve the climbing capabilities and the locomotion efficiency, a control method minimizing wheel slip has been developed in this thesis. Unlike other control strategies, the proposed method does not require the use of soil models. Independence from these models is very significant because the ability to operate on different types of soils is the main requirement for exploration missions. Moreover, our approach can be adapted to any kind of wheeled rover and the processing power needed remains relatively low, which makes online computation feasible. In rough terrain, the problem of tracking the robot's position is tedious because of the excessive variation of the ground. Further, the field of view can vary significantly between two data acquisition cycles. In this thesis, a method for probabilistically combining different types of sensors to produce a robust motion estimation for an all-terrain rover is presented. The proposed sensor fusion scheme is flexible in that it can easily accommodate any number of sensors, of any kind. In order to test the algorithm, we have chosen to use the following sensory inputs for the experiments: 3D-Odometry, inertial measurement unit (accelerometers, gyros) and visual odometry. The 3D-Odometry has been specially developed in the framework of this research. Because it accounts for ground slope discontinuities and the rover kinematics, this technique results in a reasonably precise 3D motion estimate in rough terrain. The experiments provided excellent results and proved that the use of complementary sensors increases the robustness and accuracy of the pose estimate. In particular, this work distinguishes itself from other similar research projects in the following ways: the sensor fusion is performed with more than two sensor types and sensor fusion is applied a) in rough terrain and b) to track the real 3D pose of the rover. Another result of this work is the design of a high-performance platform for conducting further research. In particular, the rover is equipped with two computers, a stereovision module, an omnidirectional vision system, an inertial measurement unit, numerous sensors and actuators and electronics for power management. Further, a set of powerful tools has been developed to speed up the process of debugging algorithms and analyzing data stored during the experiments. Finally, the modularity and portability of the system enables easy adaptation of new actuators and sensors. All these characteristics speed up the research in this field.

EPFL2005

3D position tracking for all-terrain robots

Pierre Lamon

EPFL2005