A qualitative path planner for robot navigation using human-provided maps

Abstract

A method for controlling a mobile robot using qualitative inputs in the context of an approximate map, such as one sketched by a human, is presented. By defining a desired trajectory with respect to observable landmarks, human operators can send semi-autonomous robots into areas for which a true map is not available. Waypoint planning is formulated as a quadratic optimization problem which takes advantage of the probabilistic representation of the observed environment and the uncertain human input, resulting in robot trajectories in the true environment that are qualitatively similar to those provided by the human. This paper formally presents a methodology in which waypoints are extracted from a hand-drawn sketch, and obstacle avoidance is naturally accommodated through the addition of constraints in the optimization problem. A sensitivity analysis is performed to study how map distortions, sensor constraints, and a priori knowledge of the map orientation affect the performance of the planner. Lastly, a set of user studies is presented to demonstrate the robustness of the planner to different users' sketched maps and to illustrate the efficacy of such a method for mobile robot control.

URSIM: Unique Regions for Sketch Map Interpretation and Matching

“…Shah and Campbell [18] performed navigation using a sketch map of obstacles provided by the user. They extracted waypoints from the Voronoi diagram and estimated the transformation between the sketch map and the environment using a weighted least square affine matrix.…”

SLAM auto-complete: Completing a robot map using an emergency map

“…Another idea is to consider the prior map as a topological representation of the environment. Shah and Campbell [1] use a human-provided map representing buildings to create waypoints using the Voronoi-Delaunay graph. The prior map is matched onto the robot map to estimate an affine transformation.…”

“…Center of Applied Autonomous Sensor Systems (AASS), Örebro University, Sweden. firstname.lastname@oru.se This work was funded in part by the EU H2020 project SmokeBot (ICT-23-2014 645101) and by the Swedish Knowledge Foundation under contract number 20140220 (AIR) Previous works on SLAM with prior information focused on using either a topological map depicting objects [1] or a map representing an environment with no distortions or errors, e.g. aerial maps [2]- [4] or highly accurate models of the environment [5], [6].…”

“…While the methods mentioned above can deal with outdated data, they assume that the prior is a metrically accurate map. It is not straight forward to integrate emergency maps in SLAM using approaches from the literature [1]- [6] due to the non-uniform scale.…”

Continuous learning route map for robot navigation using a growing-on-demand self-organizing neural network

“…[1][2][3] As an important kind of behavioral strategy in human navigation, route-based navigation strategy has been used in autonomous mobile robots, that is, a process of using real-time perceived environment information and selfmotion ability to reach the destination along the preset route. [4][5][6] Route serves as the basis of route-based navigation strategy, represented as a sequence of decision points, and considered as a part of the spatial knowledge. 7 The integration of different routes forms a network-like structure, called route map, which can represent complex and large-scale navigation environment.…”