佐治亚理工最新研究成果：会漂移的无人驾驶汽车_【GT校友会】

佐治亚理工学院的研究团队发明了一种有助于使无人驾驶车辆在操作极限边缘依然保持可控的新方法。该方法可以帮助未来的无人驾驶汽车在危险的道路条件下保持安全。...

目前，无人驾驶汽车可以完成大部分基本的驾驶任务，例如变道或停车，但如何避免迎面而来的车辆或是绕开路上突然出现的小动物，对于大多数驾驶系统来说仍然太复杂。当谷歌气泡形状的无人驾驶汽车在加州闲逛时，当特斯拉的Autopilot自动停车入库时，来自佐治亚理工学院的研究团队发明了一种有助于使无人驾驶车辆在操作极限边缘依然保持可控的新方法。该方法可以帮助未来的无人驾驶汽车在危险的道路条件下保持安全。

Autonomous cars can currently perform most of the basic driving tasks, like switching lanes or parking, but avoiding an oncoming vehicle or avoiding an animal on the road is still too complex for most systems. While Google's bubble-shaped car putters around in California and Tesla's Autopilot guides cars into garages, a Georgia Institute of Technology research team has devised a novel way to help keep a driverless vehicle under control as it maneuvers at the edge of its handling limits. The approach could help make self-driving cars of the future safer under hazardous road conditions.

来自佐治亚理工Daniel Guggenheim航空航天工程学院和交互计算学院的研究团队制造出一款正常大小五分之一规模的全自动拉力赛汽车，它能在90英里每小时（约145km/h）的速度下完成跳跃和漂移。进行技术实验的迷你拉力赛车携带一块配有四核处理器的主板，一个强大GPU和一块电池。每一辆车还有两个前置摄像头，一个惯性测量单元，一个GPS信号接收器，以及先进的轮速传感器。电力、导航、和计算设备是安装在一个能承受剧烈翻转的坚固铝合金外壳下。每辆车的重量约为48磅（约22kg），约3英尺长（约91cm）。

A team of researchers from Georgia Tech’s Daniel Guggenheim School of Aerospace Engineering (AE) and the School of Interactive Computing (IC) have built auto-rally, a one-fifth-scale, fully autonomous rally truck that can jump and drift at the equivalent of 90 mph. The auto-rally vehicles carry a motherboard with a quad-core processor, a potent GPU, and a battery. Each vehicle also has two forward-facing cameras, an inertial measurement unit, and a GPS receiver, along with sophisticated wheel-speed sensors. The power, navigation, and computation equipment is housed in a rugged aluminum enclosure able to withstand violent rollovers. Each vehicle weighs about 48 pounds and is about three feet long.

佐治亚理工学院研究团队解读最新无人驾驶车辆操控技术

“无人驾驶汽车应该能够处理任何状况，而不只是在高速公路上的正常情况下。我们的主要目标之一是将人类驾驶的一些技术手法注入到无人驾驶汽车的‘大脑’中去。”佐治亚理工航空航天工程学院教授Panagiotis Tsiotras这样说道，他同时也是一位在拉力赛车速控制方面的数学专家。

“An autonomous vehicle should be able to handle any condition, not just drive on the highway under normal conditions,” said Panagiotis Tsiotras, an AE professor who is an expert on the mathematics behind rally-car racing control. “One of our principal goals is to infuse some of the expert techniques of human drivers into the brains of these autonomous vehicles.”

传统的机器人车辆技术无论在正常驾驶的情况下还是在道路边缘驾驶的情况下都采用相同的控制方法，Tsiotras教授解释道。佐治亚理工的技术方法–称为模型预测路径积分控制（Model Predictive Path Integral Control，MPPI）–是专门开发，以解决涉及在极限摩擦情况下控制车辆的非线性动力学。该技术采用先进的算法和车载计算，与安装在车上的的传感装置相呼应，在保持性能的同时提高车辆的稳定性。

Traditional robotic-vehicle techniques use the same control approach whether a vehicle is driving normally or at the edge of roadway adhesion, Tsiotras explained. The Georgia Tech method – known as model predictive path integral control (MPPI) – was developed specifically to address the non-linear dynamics involved in controlling a vehicle near its friction limits. The technique uses advanced algorithms and onboard computing, in concert with installed sensing devices, to increase vehicular stability while maintaining performance.

佐治亚理工学院的研究人员使用了一种基于路径积分方法进行随机轨迹优化的能力，创造他们的MPPI控制算法，Tsiotras教授解释道。使用统计方法，研究团队集成了大量的操作相关信息，以及车辆系统的动态数据，从无数的可能性中计算出最稳定的轨迹。

The Georgia Tech researchers used a stochastic trajectory-optimization capability, based on a path-integral approach, to create their MPPI control algorithm, Theodorou explained. Using statistical methods, the team integrated large amounts of handling-related information, together with data on the dynamics of the vehicular system, to compute the most stable trajectories from myriad possibilities.

通过车载的大功率图形处理单元（GPU）的处理，MPPI控制算法不断从全球定位系统（GPS）硬件、惯性运动传感器和其他传感器方面提取样本数据。车载的硬件软件系统对大量的可能轨迹进行实时分析，并时时向车辆传达最优操控决定。当拉力赛车在测试轨道上猛烈转弯的时候，车载的GPU会使MPPI算法在1/60秒内计算出超过2500个2.5秒长的轨迹，并从中做出最优路径和动力选择。

Processed by the high-power graphics processing unit (GPU) that the vehicle carries, the MPPI control algorithm continuously samples data coming from global positioning system (GPS) hardware, inertial motion sensors, and other sensors. The onboard hardware-software system performs real-time analysis of a vast number of possible trajectories and relays optimal handling decisions to the vehicle moment by moment.

When aggressively going round a corner at the university's test track, the onboard GPU lets the MPPI algorithm sample more than 2,500, 2.5-second-long trajectories in under 1/60 of a second to figure out the best trajectory to take and the right amount of power to put down.

在本质上，MPPI方法将最优操控决定的计划和执行组合为一个高效的阶段，它被认为是首个能够实现如此艰巨计算任务的技术。在过去，最优控制数据的输入并不能做到实时处理。

In essence, the MPPI approach combines both the planning and execution of optimized handling decisions into a single highly efficient phase. It’s regarded as the first technology to carry out this computationally demanding task; in the past, optimal- control data inputs could not be processed in real time.

几十年之后，当你的自驾谷歌汽车在结冰的路面上打滑后依旧保持平衡时，你会感激佐治亚理工所做的工作。

In a few decades when your self-driving Google car stays on the road after skidding round an icy corner, you'll be grateful for Georgia Tech's work.

（该文章英文原文节选自Popular Science, 英国每日邮报、独立报等多篇媒体报道）