Computer Engineering, San Jose State University, San Jose, CA 95123, USA.
Sensors (Basel). 2018 Jun 5;18(6):1829. doi: 10.3390/s18061829.
Despite the technical success of existing assistive technologies, for example, electric wheelchairs and scooters, they are still far from effective enough in helping those in need navigate to their destinations in a hassle-free manner. In this paper, we propose to improve the safety and autonomy of navigation by designing a cutting-edge autonomous scooter, thus allowing people with mobility challenges to ambulate independently and safely in possibly unfamiliar surroundings. We focus on indoor navigation scenarios for the autonomous scooter where the current location, maps, and nearby obstacles are unknown. To achieve semi-LiDAR functionality, we leverage the gyros-based pose data to compensate the laser motion in real time and create synthetic mapping of simple environments with regular shapes and deep hallways. Laser range finders are suitable for long ranges with limited resolution. Stereo vision, on the other hand, provides 3D structural data of nearby complex objects. To achieve simultaneous fine-grained resolution and long range coverage in the mapping of cluttered and complex environments, we dynamically fuse the measurements from the stereo vision camera system, the synthetic laser scanner, and the LiDAR. We propose solutions to self-correct errors in data fusion and create a hybrid map to assist the scooter in achieving collision-free navigation in an indoor environment.
尽管现有的辅助技术（例如电动轮椅和电动滑板车）在技术上取得了成功，但它们在帮助有需要的人轻松到达目的地方面仍然远远不够有效。在本文中，我们提出通过设计一种先进的自动驾驶滑板车来提高导航的安全性和自主性，从而使行动不便的人能够在可能不熟悉的环境中独立且安全地行走。我们专注于自动驾驶滑板车的室内导航场景，其中当前位置、地图和附近障碍物是未知的。为了实现半激光雷达功能，我们利用基于陀螺仪的姿态数据实时补偿激光运动，并创建具有规则形状和深走廊的简单环境的合成映射。激光测距仪适用于远距离和有限分辨率的情况。另一方面，立体视觉提供了附近复杂物体的三维结构数据。为了在杂乱和复杂环境的映射中实现同时的精细分辨率和长距离覆盖，我们动态融合来自立体视觉相机系统、合成激光扫描仪和激光雷达的测量值。我们提出了解决数据融合中错误自校正的方法，并创建了一个混合地图，以帮助滑板车在室内环境中实现无碰撞导航。
