Automatic parking trajectory planning method based on vehicle kinematic model

Abstract

The invention relates to the field of intelligent automobiles, in particular to an automatic parking track planning method based on a vehicle kinematic model, which comprises the following steps of: receiving vehicle positioning information and target parking space position information through a combined positioning device, acquiring surrounding obstacle information through a forward camera, a millimeter-wave radar and an ultrasonic radar; and obtaining the corner point information of the parking space through the look-around sensing module. A vehicle carries out parking path planning according to the current posture, parking space information and lane width information, according to different parking space types, parking space information constraints, road width constraints and a vehicle kinematic model, a one-time vertical parking planning method, a multi-section type vertical parking planning method and a pouring type parallel parking planning method are designed, and after planning succeeds, the vehicle can be parked according to the parking space types, the parking space information constraints, the road width constraints and the vehicle kinematic model. And the track control point is sent to a vehicle-mounted controller, and the vehicle-mounted controller controls the vehicle to be parked in the target parking space. Compared with the prior art, the method has the advantages of being high in parameter adjusting capacity, good in algorithm real-time performance and the like.

Description

technical field [0001] The invention relates to the field of smart cars, in particular to an automatic parking trajectory planning method based on a vehicle kinematics model. Background technique [0002] At present, there is a huge unmet demand in China's parking market, and there are strong pain points such as "difficult to park and difficult to pick up" in parking. In addition, according to the statistics of traffic accident database and accident statistics of insurance companies, accidents caused by parking accounted for 44% of various accidents, of which about 1 / 2 to 3 / 4 parking collisions are caused by reversing. The autonomous parking system can not only improve the safety rate of parking, but also improve the utilization rate of parking spaces in the parking lot. Almost all currently commercialized parking assistance systems are based on various sensors and infrastructure, and have passive safety assistance functions, such as warning the driver of the presence of ob...

AI Technical Summary

Problems solved by technology

The search-based A* algorithm is very suitable for unmanned vehicle path planning on unstructured roads, and has its unique algorithm advantages in parking conditions. The path planning method for unmanned vehicles is more likely to fail in search, and it is not conducive to the trajectory tracking of the control module, so that the expected parking accuracy cannot be achieved. In addition, the search algorithm needs to perform high-precision discretization of the surrounding environment map, which greatly affects the algorithm. Real-time, not applicable in parking environment

Images