|Abstract:||High-accuracy positioning and the different techniques that can be used to achieve it, such as Real-Time Kinematics (RTK), Precise Point Positioning (PPP), Wide-RTK or PPP-RTK have been trending topic in the last decade due to the huge number of potential applications that take benefit from these technologies, now and for sure in the future, and thanks also to the novel business opportunities that they bring. Examples of classical user applications of these high-accuracy techniques are high-precision agriculture, mapping, survey and many others. Today, the irruption of new markets in the near horizon imposes new challenges in order to meet their requirements. It happens with the use of GNSS for autonomous driving, unmanned autonomous vehicles or automatic fleet management, to name just a few examples. They raise the need to adapt the existing technologies, evolve them or even go for ground-breaking alternative solutions. The use of GNSS for autonomous driving, or even for autonomous applications in general, brings practically the main difficult challenges, when talking about high and safe accuracy positioning, all together at the same time: high-dynamics conditions up to almost 200 km/h, low bandwidth available for GNSS data though terrestrial channels, dead-reckoning robustness, fast-convergence at initialization and after GNSS outages, cost-effective infrastructure constraints in terms of for example the number of reference stations, degraded environments in terms of GNSS satellites in view or multipath conditions and the necessity to maintain a safe state under any circumstance or hazard. New high demanding problems claim for the most advanced engineering solutions, supported by well-based mathematical background, and state-of-the-art, or even disruptive, innovative techniques. In the context described in the previous paragraph, this paper will present a solution based on the use of advanced PPP algorithms, integrated with low-cost inertial sensors (IMUs and odometers), running in a mass-market HW/SW platform, compatible with mass-market GNSS receivers and low-cost antennas. Our solution is able to achieve the stringent integrity requirements from safety-critical automotive applications, based on an integrity concept driven by a solid and methodical safety assessment. At the same time, it provides a high accuracy with a fast convergence, even under the frequent GNSS outages found in many conditions. The objective of this paper is to provide an overview of GMV’s solution for autonomous applications, which is based on magicPPP algorithms (www.magicgnss.gmv.com). The paper will describe the high-level architecture of the solution and different algorithms and techniques which support it. In the last months, an extensive experimentation campaign based mainly on real data driving tests and some simulations have been performed. The results and statistics of this experimentation campaign will be presented and analysed.|
Proceedings of the 32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019)
September 16 - 20, 2019
Hyatt Regency Miami
|Pages:||1707 - 1737|
|Cite this article:||
Calle, David, Carbonell, Enrique, Carretero, José Luis, Chamorro, Adrián, Durán, Jorge, Mezzera, Cecilia, Navarro, Pedro, Rodríguez, Irma, "Cutting-edge Technical Solutions for the Next Generation of Autonomous Vehicle," Proceedings of the 32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019), Miami, Florida, September 2019, pp. 1707-1737.
ION Members/Non-Members: 1 Download Credit