Lane pickup algorithm for autopilot high-precision map

Abstract

The invention discloses a lane pickup algorithm of automatic driving high-precision map. The algorithm comprises the following steps: 1. If the positioning precision of the positioning module drops, the lane pickup algorithm enters an emergency retrieval state; 2, judging whether to enter a local search or not, and if so, entering that step 6; 3, obtaining that position range of the vehicle and the street ID within the position range, then entering the step 4, searching the range of the vehicle, searching the set of the streets where the vehicle may be located, and entering the step 5, searching roughly to find the set of the streets where the vehicle may be currently located; If not, entering step 4, expanding the search range and searching; If so, proceeding to step 6 for a detailed search to obtain the first element in the set of lanes in which the vehicle may be located, that is, the lane most likely to be located. The invention can quickly and efficiently retrieve the lane-relatedinformation of the current vehicle from the high-precision map database.

Description

technical field [0001] The invention relates to the technical field of automatic driving, in particular to a lane picking algorithm for a high-precision map for automatic driving. Background technique [0002] With the continuous development of autonomous driving technology and the continuous clarification of commercial landing scenarios, high-precision maps and their applications are also playing an increasingly important role, serving perception, positioning, planning control, decision-making, and human-computer interaction. Modules of the autonomous driving system. From a technical point of view, compared with traditional maps, while providing road network topological relationships, high-precision maps have richer road information at the lane level, which not only provides important prior data support for automatic driving systems, but also guarantees Driving safety provides strong system redundancy. From the perspective of commercial implementation, the various sensors...

AI Technical Summary

Problems solved by technology

Compared with the traditional navigation map that is accurate to the street level, the high-precision map for autonomous driving can be accurate to the lane level and has richer information, but it also increases the difficulty of data retrieval

Images