Research On Obstacle Avoidance System And Path Planning Of Unmanned Ground Vehicle Based On PX4

Unmanned ground vehicle obstacle avoidance and path planning means that the unmanned ground vehicle encounters obstacles during automatic cruise and can automatically avoid and quickly return to the preset route.At present unmanned vehicles play an irreplaceable role in both military and daily life,automatic obstacle avoidance is an essential function of unmanned vehicles.In this paper the open source flight control Pixhawk is used as the motion controller and the PX4 Rover software control system used by the unmanned ground vehicle is redeveloped to realize the obstacle avoidance and path planning function in the autonomous cruise mode.This paper first analyzes the implementation of the automatic cruise function of the unmanned vehicle.This function is the basis for obstacle avoidance and path planning.Through multi-sensor data fusion accurate state information during the operation of the unmanned vehicle is obtained.Based on the bicy'cle model the unmanned vehicle is modeled and analyzed.According to the established model and the cascade PID algorithm the unmanned vehicle running path is dynamically adjusted to realize the automatic cruising function of the unmanned vehicle.After implementing the auto-cruise function another problem that needs to be solved is the obstacle detection problem.To solve this problem multiple ultrasonic waves are used to obtain obstacle information around the unmanned vehicle.Because the Pixhawk controller does not support multiple external ultrasonic sensors a data acquisition module is designed based on the Arduino Mega2560 to collect multiple ultrasonic data and send the processed data to the Pixhawk controller through the serial port.Because Pixhawk uses the communication mechanism of uORB the data sent by the data acquisition module to the Pixhawk through the serial port cannot be directly used.Therefore the ultrasonic data is published and subscribed by means of a custom theme.After detecting obstacles how to achieve fast and accurate obstacle avoidance is another problem that needs to be solved in this paper.According to the distance of the obstacle the sensor detection area is divided into the deceleration area the obstacle avoidance area and the stop area and the different orientations of the obstacle are determined according to the detection result of the sensor.Different obstacle avoidance strategies are adopted for obstacles in different orientations.Aiming at how to return to the original route after obstacle avoidance combined with the cascade PID algorithm in path planning a rapid adjustment strategy is formulated to optimize the path.Finally using the obstacle avoidance and path planning strategy designed in this paper the experimental verification is carried out in the real scene.The experiment shows that the obstacle avoidance strategy of this paper can achieve better obstacle avoidance effect under various obstacle environments.