Real-time footstep planning using a geometric approach
Authors:
- 1 May 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1782-1787
Abstract
To this date, many footstep planning systems rely on external sensors for mapping and traversability analysis or on computationally expensive algorithms that do not allow for real-time calculations. In this paper, we present an approach that analyzes the environment in the vicinity of the robot with an onboard RGBD camera while computing local footstep plans in real time. We achieve this by combining the advantages of grid-based height maps, fast planar region segmentation, and a systematic local footstep search to a local goal point. Using a single CPU core, a full mapping and planning cycle only takes 18 ms on average and thus presents an important step to autonomous humanoid robots in dynamic environments that only rely on their onboard hardware.Keywords
This publication has 11 references indexed in Scilit:
- Continuous humanoid locomotion over uneven terrain using stereo fusion2015
- Computing Large Convex Regions of Obstacle-Free Space Through Semidefinite Programming2015
- Supervised footstep planning for humanoid robots in rough terrain tasks using a black box walking controller2014
- Footstep planning on uneven terrain with mixed-integer convex optimization2014
- Integrated perception, mapping, and footstep planning for humanoid navigation among 3D obstacles2013
- Fast and accurate plane segmentation in depth maps for indoor scenes2012
- Adaptive level-of-detail planning for efficient humanoid navigation2012
- Real-time replanning using 3D environment for humanoid robot2011
- An adaptive action model for legged navigation planning2007
- Vision-guided humanoid footstep planning for dynamic environments2006