Multi-sensor fused unmanned vehicle detection obstacle avoidance system and obstacle avoidance method

Abstract

The invention provides a multi-sensor fused unmanned vehicle detection obstacle avoidance system and obstacle avoidance method. The obstacle avoidance system comprises a plurality of single-line laserradars, double CCD cameras, microwave radars, front and rear blind area ultrasonic sensor groups and a three-core controller based on ARM + FPGA + NUC. A plurality of single-line laser radar signalsare processed by the NUC microcomputer, binocular vision graphic data of the CCD camera are jointly processed by the ARM + FPGA controller, functions of blind area detection and obstacle avoidance, ahuman-computer interface, path planning, online output and the like are independently completed by the STM32F767, and the ARM + FPGA controller outputs control signals through decoding to accurately control the direct-current brushless servo motor, and drives an unmanned vehicle to run. According to the invention, the unmanned vehicle can discover obstacles in a complex environment in an all-weather and farther manner and quickly realize effective obstacle avoidance, so the safety and stability of the unmanned vehicle during high-speed driving are improved.

Description

technical field [0001] The invention belongs to the technical field of unmanned driving, and in particular relates to a multi-sensor fusion unmanned vehicle detection and obstacle avoidance system and an obstacle avoidance method. Background technique [0002] With the rapid economic development, cars have become an increasingly important part of people's lives. The negligence of drivers will lead to many accidents. Therefore, car manufacturers concentrate on designing systems that can ensure the safety of cars. Safety is one of the main factors driving the growth of demand for driverless cars; Traffic congestion makes driving not so beautiful. Artificial intelligence driverless cars can completely solve traffic congestion and other problems. In addition, poor air conditions are also a "catalyst" for the promotion of driverless cars. [0003] A driverless car is a smart car that senses the road environment through an on-board sensor system, automatically plans a driving rou...

AI Technical Summary

Problems solved by technology

[0007] (1) Since the unmanned vehicle is disturbed by unstable factors in the surrounding environment, the controller based on the single-chip microcomputer has poor anti-interference ability, and abnormalities often occur, causing the unmanned vehicle to lose control;

[0008] (2) The existing unmanned vehicles all use low-level DSP and ARM series chips, and the maximum operating frequency is only about 100 MHz, which cannot meet the fast calculation of complex data of unmanned vehicles;

[0009] (3) Affected by the performance of the PC of the unmanned vehicle, the data collected by the sensor of the unmanned vehicle cannot be quickly calculated and stored;

[0010] (4) The data acquired by the single-line laser radar is 2D data, and information such as the height of the target cannot be distinguished. Some small objects will be ignored and eventually become obstacles. Single single-line laser radar sensor navigation has become a bottleneck in the vehicle field;

[0011] (5) A single single-line laser radar cannot obtain road surface information, and it needs to cooperate with other sensors to read and judge ground information;

[0012] (6) Although multi-line laser radar can realize 2.5D or 3D data, can judge the height of obstacles, process ground information, etc., but the price is relatively expensive. Large-scale promotion and use;

[0013] (7) A single single-line laser radar cannot detect information such as corners and road cliffs, and it needs to be used with other sensors to read the surrounding obstacle signals or locate the sensor signs;

[0014] (8) The existing unmanned vehicles basically only consider forward detection and obstacle avoidance, and do not consider the obstacle information in the rear. Realize accelerated avoidance;

[0015] (9) Based on a single single-line lidar unmanned vehicle, there is a detection blind spot at the moment of start-up. Once an obstacle is in the blind spot, traffic accidents are prone to occur;

[0016] (10) Unmanned vehicles based on a single single-line lidar will also have detection blind spots during actual driving. Once obstacles enter the blind spot during movement, traffic accidents will also occur;

[0017] (11) The unmanned vehicle based on the single-line lidar has a slow speed of image acquisition of the road ahead, which affects the rapid progress of the unmanned vehicle;

[0018] (12) During long-distance driving, unmanned vehicles based on single-line lidar have poor recognition of the surrounding environment and cannot achieve accurate positioning;

[0019] (13) In the regular traffic, there are various traffic signs on the ground of the driving path of the unmanned vehicle, but the single-line lidar cannot be recognized, and the auxiliary navigation when the unmanned vehicle is moving rapidly is lost;

[0020] (14) In regular traffic, there are signs such as traffic lights in the air on the driving path of unmanned vehicles, but the single-line lidar cannot be identified, which weakens the safety of unmanned vehicles when they are moving fast;

[0021] (15) Affected by the price and performance of lidar, the detection range of generally cost-effective lidar is less than 100 meters. This distance is not conducive to the judgment of unmanned vehicles moving quickly.

[0023] (1) Affected by the defects of monocular vision itself, the sensor detection system of unmanned vehicles has improved compared with single-line lidar, but this distance is not conducive to high-speed driving of unmanned vehicles;

[0024] (2) CCD-based monocular visual obstacle discrimination requires a feature library with a large data capacity. Once an object does not have a feature library data to match it, obstacles will not be able to be identified, and thus it will not be possible to accurately estimate these targets. The distance is not conducive to the high-speed driving of unmanned vehicles;

[0025] (3) Whether it is a single-line laser radar, a multi-line laser radar or a camera, it is very sensitive to weather with rain, fog, dust, and smog. The performance of weather, lidar and video signals will be greatly reduced, which will have a greater impact on the safety of unmanned vehicles;

[0026] (4) Whether it is single-line lidar, multi-line lidar or camera is very sensitive to strong light weather, strong sunlight can sometimes greatly reduce the performance of lidar and camera, and sometimes there is no signal output, which is harmful to no one. significant impact on vehicle safety

Images