当前汽车技术发展主要是面向新四化，即电动化、智能化、网联化、共享化。无人驾驶技术是汽车技术新四化中重要的组成部分，它在增强高速公路安全、缓解交通拥堵、减少空气污染等领域带来颠覆性的改善。同时无人驾驶技术也将促进智慧交通技术和智慧城市的发展。
在无人驾驶技术组成单元中，环境感知系统是至关重要的一环，通过多种传感器采集无人驾驶车辆所需的自身姿态和周边环境信息，是无人车安全性和智能性的保障。在环境感知系统部分主要配置基础是视觉传感器和激光雷达，另外还有GPS/IMU、雷达和声呐等。激光雷达可以检测目标物体的方位和距离的三维信息，并可通过点云来描述道路情况的3D环境模型。另一个方面，在现有的产品自动驾驶车辆中，因为激光雷达成本高，价格相对低廉的摄像机被赋予了更多的感知任务。视觉感知技术包括基本的车道标志线检测、交通标志以及交通信号灯检测，到交通环境中物体的识别与跟踪、车辆同步定位建图和导航等。
本课程主要围绕无人驾驶中视觉感知和激光雷达技术展开讲授，并结合智能车平台，让学生在短期内了解和掌握无人驾驶环境感知系统的原理、组成和操作过程等内容，并且具有基本的调试动手能力。
At present, the development of automobile technology is oriented to electrification, intelligentization, networking and sharing. Unmanned vehicle technology is an important part of the new modernization of automobile technology, which brings great improvements in areas such as enhancing highway safety, easing traffic congestion and reducing air pollution. At the same time, unmanned vehicle technology will also promote the development of intelligent transportation technology and smart city.
In the component unit of unmanned vehicle driving technology, the environmental perception system is a crucial part. The self-posture and surrounding environment information required by the unmanned vehicle can be collected through a variety of sensors, which guarantees the safety and intelligence of the unmanned vehicle. In the environmental sensing system, the main sensors are camera sensor, lidar, GPS/IMU, radar and sonar. Lidar can detect the 3D information of the target object's orientation and distance, and can describe the 3D environmental model of road conditions through point cloud. On the other hand, in existing production of autonomous vehicles, cameras are given more perceptual tasks. Visual perception technology includes line detection, traffic sign and traffic signal detection, identification and tracking of objects, simultaneous localization and mapping.
This course mainly focuses on the visual perception and lidar technology in unmanned vehicle. Combined with the intelligent vehicle platform, it enables students to understand and master the principle, composition and operation process of the unmanned vehicle perception system, and has debugging and hands-on ability.
|
