Obstacle Avoidance Flight Research Articles

Ro2En: Robust Neural Environment Encoder for Domain Generalization of Fast Motion Planning

This paper discusses a new issue named domain generalization of fast motion planning in 3D environments, which benefits agility-required robot applications such as autonomous driving and uncrewed aerial vehicle obstacle avoidance flight. The existing work shows that conventional spatial search-based planning algorithms cannot meet the real-time requirement due to high time costs. The end-to-end neural network-based methods achieve an excellent balance between performance and planning speed in the seen environments, but are hard to transfer to new scenarios. To overcome this limitation, we propose a novel Robust Environment Encoder (Ro2En) approach to domain generalization of fast motion planning. Specifically, by demonstrating the reconstructed environment, we find that the previous environment encoder cannot encode the volume information properly, i.e., a volume collapse ensues, which leads to noisy environment modeling. Inspired by this observation, a dual-task auto-encoder is developed. It can not only reconstruct the point cloud of the obstacles, but also align their geometric centers. Experiment results showed that in the new scenarios, Ro2En outperformed previous state-of-the-art conventional and neural alternatives with a much smaller performance variation.

Analysis of collision avoidance in honeybee flight.

Insects are excellent at flying in dense vegetation and navigating through other complex spatial environments. This study investigates the strategies used by honeybees (Apis mellifera) to avoid collisions with an obstacle encountered frontally during flight. Bees were trained to fly through a tunnel that contained a solitary vertically oriented cylindrical obstacle placed along the midline. Flight trajectories of bees were recorded for six conditions in which the diameter of the obstructing cylinder was systematically varied from 25 mm to 160 mm. Analysis of salient events during the bees' flight, such as the deceleration before the obstacle, and the initiation of the deviation in flight path to avoid collisions, revealed a strategy for obstacle avoidance that is based on the relative retinal expansion velocity generated by the obstacle when the bee is on a collision course. We find that a quantitative model, featuring a controller that extracts specific visual cues from the frontal visual field, provides an accurate characterization of the geometry and the dynamics of the manoeuvres adopted by honeybees to avoid collisions. This study paves the way for the design of unmanned aerial systems, by identifying the visual cues that are used by honeybees for performing robust obstacle avoidance flight.

Autonomous aerial obstacle avoidance using LiDAR sensor fusion.

The obstacle avoidance problem of unmanned aerial vehicle (UAV) mainly refers to the design of a method that can safely reach the target point from the starting point in an unknown flight environment. In this paper, we mainly propose an obstacle avoidance method composed of three modules: environment perception, algorithm obstacle avoidance and motion control. Our method realizes the function of reasonable and safe obstacle avoidance of UAV in low-altitude complex environments. Firstly, we use the light detection and ranging (LiDAR) sensor to perceive obstacles around the environment. Next, the sensor data is processed by the vector field histogram (VFH) algorithm to output the desired speed of drone flight. Finally, the expected speed value is sent to the quadrotor flight control and realizes autonomous obstacle avoidance flight of the drone. We verify the effectiveness and feasibility of the proposed method in 3D simulation environment.

The Control Method of Autonomous Flight Avoidance Barriers of UAVs in Confined Environments

This paper proposes an improved 3D-Vector Field Histogram (3D-VFH) algorithm for autonomous flight and local obstacle avoidance of multi-rotor unmanned aerial vehicles (UAVs) in a confined environment. Firstly, the method employs a target point coordinate system based on polar coordinates to convert the point cloud data, considering that long-range point cloud information has no effect on local obstacle avoidance by UAVs. This enables UAVs to effectively utilize obstacle information for obstacle avoidance and improves the real-time performance of the algorithm. Secondly, a sliding window algorithm is used to estimate the optimal flight path of the UAV and implement obstacle avoidance control, thereby maintaining the attitude stability of the UAV during obstacle avoidance flight. Finally, experimental analysis is conducted, and the results show that the UAV has good attitude stability during obstacle avoidance flight, can autonomously follow the expected trajectory, and can avoid dynamic obstacles, achieving precise obstacle avoidance.

Autonomous Navigation and Obstacle Avoidance for Small VTOL UAV in Unknown Environments

This paper takes autonomous exploration in unknown environments on a small co-axial twin-rotor unmanned aerial vehicle (UAV) platform as the task. The study of the fully autonomous positioning in unknown environments and navigation system without global navigation satellite system (GNSS) and other auxiliary positioning means is carried out. Algorithms that are based on the machine vision/proximity detection/inertial measurement unit, namely the combined navigation algorithm and indoor simultaneous location and mapping (SLAM) algorithm, are not only designed theoretically but also realized and verified in real surroundings. Additionally, obstacle detection, the decision-making of avoidance motion and motion planning methods such as Octree are also proposed, which are characterized by randomness and symmetry. The demonstration of the positioning and navigation system in the unknown environment and the verification of the indoor obstacle-avoidance flight were both completed through building an autonomous navigation and obstacle avoidance simulation system.

A high-accuracy calibration method for fusion systems of millimeter-wave radar and camera

Multi-sensor information fusion is widely used in the field of unmanned aerial vehicles obstacle avoidance flight, particularly in millimeter-wave (MMW) radar and camera fusion systems. Calibration accuracy plays a crucial role in fusion systems. The low-angle measurement accuracy of the MMW radar usually causes large calibration errors. To reduce calibration errors, a high-accuracy calibration method based on a region of interest (ROI) and an artificial potential field was proposed in this paper. The ROI was selected based on the initial calibration information and the MMW radar’s angle measurement error range from the image. An artificial potential field was established using the pixels of the ROI. Two moving points were set at the left and right ends of the ROI initially. The potential forces of the two moving points are different because the pixels of the obstacle and the background are different in the image. The two moving points were iteratively moved towards each other according to the force until their distance was less than the iteration step. The new calibration point is located in the middle of the final position of the two moving points. In contrast to the existing calibration methods, the proposed method avoids the limitations of low angle measurement accuracy by using image pixels. The experimental results show that the calibration errors decrease by 83.95% and 75.79%, which is significantly improved compared to the traditional methods and indicates the efficiency of the proposed method.

Bumblebees display characteristics of active vision during robust obstacle avoidance flight.

Insects are remarkable flyers and capable of navigating through highly cluttered environments. We tracked the head and thorax of bumblebees freely flying in a tunnel containing vertically oriented obstacles to uncover the sensorimotor strategies used for obstacle detection and collision avoidance. Bumblebees presented all the characteristics of active vision during flight by stabilizing their head relative to the external environment and maintained close alignment between their gaze and flightpath. Head stabilization increased motion contrast of nearby features against the background to enable obstacle detection. As bees approached obstacles, they appeared to modulate avoidance responses based on the relative retinal expansion velocity (RREV) of obstacles and their maximum evasion acceleration was linearly related to RREVmax. Finally, bees prevented collisions through rapid roll manoeuvres implemented by their thorax. Overall, the combination of visuo-motor strategies of bumblebees highlights elegant solutions developed by insects for visually guided flight through cluttered environments.

Laser-scanning rangefinder fixed on a gimbal with two degrees of freedom

A laser-scanning rangefinder mounted on a gimbal with two degrees of freedom is presented. The rangefinder can be used as part of a navigation system of an unmanned aerial vehicle to avoid obstacles or prevent collisions. Compared to stereo cameras, the device requires significantly less computing resources and is less dependent on lighting conditions. Compared to integrated lidars, the cost of the device is by an order lower. А model of the device was developed and an obstacle avoidance flight was simulated in the Gazebo simulator. The PX4/Avoidance software was used as an autopilot. As a result of a model experiment, we found that a scanning laser rangefinder can provide autonomous navigation with obstacle avoidance.

Insulator detection and recognition of explosion based on convolutional neural networks

Unmanned aerial vehicles (UAVs) equipped with high definition (HD) cameras can obtain a large number of detailed inspection images. The insulator is an indispensable component in the transmission lines. Detecting insulator in image video quickly and accurately can provide a reliable basis for the ranging and the obstacle avoidance flight of UAV close to the tower and transmission line. At the same time, the insulator is a serious threat to the safety of the power grid due to the multiple faults of the insulator, and the computer technology should be fully utilized to diagnose the fault. Detection of the insulator images with the complex aerial background is implemented by constructing a convolutional neural network (CNN), which has the classic architecture of five modules of convolution and pooling, two modules of fully connected layers. In this paper, we propose a recognition algorithm for explosion fault based on saliency detection, which uses the trained network model to extract the features. Then, we put the saliency maps into a self-organizing feature map (SOM) network and build the mathematical module via super pixel segmentation, contour detection and other image processing methods. The test shows that the algorithm can reduce the error that may be caused by manual analysis. It also demonstrates that the detection of the insulator and the recognition of explosion fault can effectively improve the efficiency and intelligence level.

A simplified virtual obstacle avoidance approach for aircraft altitude bounded flight maneuvers

Avoiding collision with obstacles and terrain has always been a critical issue for aircraft during low altitude flight landing and approach maneuvers, especially in airports located in mountainous areas. In low altitude flight, it is desirable to find a safe trajectory that is compatible with the aircraft capabilities and at the same time optimizes flight efficiencies. Safety is a fundamental concern when in terrain following (TF) and/or terrain avoidance (TA) flight modes, and aircraft design for these types of missions emphasizes the inclusion of terrain and obstacle avoidance flight (TOAF) capabilities.

Lift distribution of washout twist morphing MAV wing

Micro Air Vehicle, or also commonly known as MAV, is a miniature aircraft that has been gaining interest in the industry. MAV is defined as a flying platform with 15cm wingspan and operates at a speed of around 10m/s. Recently, MAV has been exposed with the latest development and link towards the biologically-inspired designs such as morphing wing. Twist morphing wing is one of the latest MAV wing design developments. The application of Twist Morphing (TM) on MAV wing has been previously known to produce better aerodynamic performance. Previous study in washin TM wing has shown a promising possibility of generating higher lift force. Despite the benevolent performance exhibited by the washin TM wing, the lift distribution for the washout type of TM MAV is relatively unknown and still open to be explored. This is probably due to the lack of experimental test rig to produce the washout twist morphing motion on the MAV wing. Therefore, this research aims to produce a special test rig for washout TM wing that is compatible for wind tunnel experimental testing. By using the special test rig, the experimental investigation on the lift performance of washout TM MAV wing can be done. Based on the wing deformation results, it clearly shows that the proposed test rig is capable to produce up to 19.5mm tip deflection at the morphing point, which is also resulting in a significant morphing motion. Higher morphing force induces larger morphing motion. Based on the lift distribution results, they show that the morphing motion has significantly affected the overall lift distribution on the MAV wing. The morphing motion on TM wing has produced at least 17.6% and 5.33% lower CL and CLmax magnitude, respectively, with the membrane wing especially at the pre-stall region. However, the TM wing is still able to maintain the stall angle similar to the baseline wing at αstall= 31°. By maintaining high αstall value with lower CL and CLmax magnitude, TM wing produces more agility for the MAV maneuverability that will be useful for indoor mission or obstacle avoidance flight.

Investigation of acoustic gaze strategy by Pipistrellus abramus and Rhinolophus ferrumequinum Nippon during obstacle avoidance flight

We investigated the acoustic gaze (angle between pulse and flight directions) and the beam width of echolocation pulses emitted by bats during obstacle avoidance flight in a chamber (7×3×2 m). Echolocation pulses were recorded by a telemetry microphone mounted on the bat’s back and a horizontal 20-ch microphone array set up in the chamber. Flight path measurements were conducted using two high-speed video cameras. While the bat showed a circular flight avoiding the surrounding walls, the acoustic gaze showed significant linear correlation to the angular velocity of the flight. This means that the bat adjusted the pulse direction to precede its own flight direction. The correlation coefficient increased with complexity of obstacle conditions. We compared changes in acoustic gaze between FM bats (mean beam width: ±50°) and CF-FM bats (mean beam width: ±22°). We found that FM bats smoothly shifted the acoustic gaze within 10° during the flight whereas the CF-FM bats frequently shifted the acoustic gaze within 25°. These results indicate that the shifting acoustic gaze by CF-FM bats compensates their own narrow beam width. Both bat species may keep approximately ±50° of their echolocation sights in order to sense the space for moving forward safely.