Reactive Navigation in Crowds for Non-Holonomic Robots With Convex Bounding Shape

Abstract

This paper describes a novel method for nonholonomic robots of convex shape to avoid imminent collisions with moving obstacles. The method's purpose is to assist navigation in crowds by correcting steering from the robot's path planner or driver. We evaluate its performance using a custom simulator which replicates real crowd movements and corresponding metrics which quantify the agent's efficiency and the robot's impact on the crowd and count collisions. We implement and evaluate the method on the standing wheelchair Qolo. In our experiments, it drives in autonomous mode using on-board sensing (LiDAR, RGB-D camera and a system to track pedestrians) and avoids collisions with up to five pedestrians and passes through a door.

GPU Accelerated Batch Multi-Convex Trajectory Optimization for a Rectangular Holonomic Mobile Robot

“…As a result, one needs to adopt techniques like sequential quadratic programming (SQP) [6], [8] accessed via frameworks like ACADO [9]. In the context of reactive onestep planning, works like [2], [10] have shown good results for navigating rectangular robots in dynamic environments. However, this setting is slightly orthogonal to our multi-step trajectory optimization set-up that performs planning over long horizons (around 30s).…”

“…E. Analysis and Batch Solution Analysing k+1 ξ 1,l : Optimization problem ( 13) is an equality constrained QP problem with same structure as (2) where…”

“…• Success-Rate: Total evaluated problems (20 ) in benchmark divided by successful runs with no-collisions. • Tracking Error: defined as (x l (t) − x des (t)) 2 + (y l (t) − y des (t)) 2 . The desired trajectory is chosen as a straight line with a constant velocity.…”

“…For example, even if the base footprint of the robot is circular when an external object (e.g., racks in warehouses) is mounted on it, the footprint invariably takes an elongated rectangular shape. A straightforward workaround is to approximate the footprint through a combination of circles [2] (see Fig. 1(a) ).…”

Unfreezing Social Navigation: Dynamical Systems based Compliance for Contact Control in Robot Navigation

“…Accepted Version: 2022 IEEE International Conference on Robotics and Automation (ICRA), May. [23][24][25][26][27]2022 -Philadelphia (PA), USA. ©2022 IEEE.…”