RGB–D terrain perception and dense mapping for legged robots

Abstract

This paper addresses the issues of unstructured terrain modeling for the purpose of navigation with legged robots. We present an improved elevation grid concept adopted to the specific requirements of a small legged robot with limited perceptual capabilities. We propose an extension of the elevation grid update mechanism by incorporating a formal treatment of the spatial uncertainty. Moreover, this paper presents uncertainty models for a structured light RGB-D sensor and a stereo vision camera used to produce a dense depth map. The model for the uncertainty of the stereo vision camera is based on uncertainty propagation from calibration, through undistortion and rectification algorithms, allowing calculation of the uncertainty of measured 3D point coordinates. The proposed uncertainty models were used for the construction of a terrain elevation map using the Videre Design STOC stereo vision camera and Kinect-like range sensors. We provide experimental verification of the proposed mapping method, and a comparison with another recently published terrain mapping method for walking robots.

D* Extra Lite: A Dynamic A* With Search–Tree Cutting and Frontier–Gap Repairing

“…Currently, we are working on the application of D* Extra Lite for path planning in environments with moving objects using the time-layered search-space architecture presented by Przybylski and Siemiątkowska (2012). In the future we plan to test D* Extra Lite in various domains, especially in hierarchical planning with the use of semantic maps that comprise topological and metrical information (Przybylski et al, 2015) combined with grid-based maps (Belter et al, 2016).…”

Registration of RGB-D Images: Components of the System

“…Istnieje szereg potencjalnych aplikacji rejestracji w robotyce [24], wśród których najważniejszymi są: budowanie trójwymiarowej mapy otoczenia (np. [2]) oraz generowanie trójwymiarowych modeli obiektów (np. [15]) na potrzeby ich późniejszego wykorzystania podczas rozpoznawania obiektów znajdujących się na scenie.…”

Single-shot Foothold Selection and Constraint Evaluation for Quadruped Locomotion

“…To efficiently navigate in a more extreme environment, perception systems are required to detect high obstacles so the robot can avoid them while walking [3]. A full 3D model of the environment can be obtained using terrain mapping methods, such as OctoMap [4], Normal Distribution Transform Occupancy Maps (NDT-OM) [5], or elevation map [6], [7]. Among all of the above, elevation *This research was supported by EU Horizon 2020 project THING.…”