Collision avoidance system and method for unmanned aerial vehicles

A sensor-agnostic and vehicle-agnostic system using multiple sensors and processors to identify and navigate around obstacles, addresses the challenge of UAV collision prevention, enhancing navigation safety and effectiveness in diverse environments.

JP7593739B2Active Publication Date: 2024-12-03AURORA FLIGHT SCIENCES CORP
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2019536047
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2017-01-06
Filing Date
2018-01-05
Publication Date
2024-12-03
Estimated Expiration
2038-01-05

AI Technical Summary

Technical Problem

Existing unmanned aerial vehicles (UAVs) face challenges in navigating around stationary and moving obstacles due to inadequate obstacle detection systems, particularly in crowded environments, leading to collisions, and current GPS-based solutions are insufficient for recognizing all objects and vary in accuracy across terrains.

Method used

A sensor-agnostic and vehicle-agnostic system using multiple sensors, including acoustic and visual, to identify obstacles within the aircraft's field of view, with a processor determining safe navigation commands by comparing pilot inputs to control commands, and overriding unsafe pilot instructions to prevent collisions.

Benefits of technology

Enhances the ability of UAVs to navigate around obstacles effectively, reducing collision risks and improving safety without requiring modifications to existing flight control systems, suitable for various environments and vehicles.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007593739000001
    Figure 0007593739000001
  • Figure 0007593739000002
    Figure 0007593739000002
  • Figure 0007593739000003
    Figure 0007593739000003
Patent Text Reader

Abstract

An obstacle avoidance system for a vehicle is provided. The system may include a communication device, a plurality of sensors configured to detect collision threats within a predetermined distance of the vehicle, and a processor. The processor may be communicatively coupled to the communication device and the plurality of sensors and configured to receive navigation commands communicated to a control system via the communication device. The processor may also receive obstacle data from at least one of the plurality of sensors reflecting a location of an obstacle. Using the obstacle data, the processor identifies a direction to avoid the obstacle. In response, the processor outputs a command to the control system via the communication device to move the vehicle toward the flight direction. Using the obstacle data, the processor may further perform a landing assistance module, a three-zone collision protection function with pilot priority, and / or a target filtering function. [Selection diagram] None
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit under 35 USC § 119(e) of U.S. Provisional Patent Application No. 62 / 443,087, filed January 6, 2017, the contents of which are incorporated herein by reference.

[0002] The present disclosure relates generally to autonomous vehicle navigation, and more particularly to systems, methods, and techniques for detecting and automatically navigating around stationary and / or moving objects. The present disclosure also relates to sensor-agnostic and vehicle-agnostic systems, methods, and techniques for detecting and automatically navigating around objects. [Background technology]

[0003] Unmanned aerial vehicle (UAV) technology has proven to be a useful tool for mission profiles including intelligence, surveillance, reconnaissance, and payload delivery. In situations such as low-altitude urban reconnaissance, UAVs, such as micro air vehicles ("MAVs"), may encounter both large and small obstacles that may be fixed or moving and whose locations are not known a priori. Furthermore, because UAVs and MAVs tend to fly in constrained and crowded environments, they may bump into or collide with objects. Furthermore, UAVs and MAVs are generally less expensive than traditional air vehicles and therefore more prevalent, and are often utilized by less skilled pilots, which may cause collisions. Existing technologies for preventing UAVs and MAVs from colliding with objects and other obstacles, such as global positioning systems ("GPS"), are generally inadequate because many objects are not visible via GPS devices and because, depending on the terrain, GPS accuracy performance varies significantly across environments.

[0004] Thus, there is a continuing need for improved autonomous vehicle navigation and obstacle avoidance systems that can react to changing and unknown obstacles in a crowded navigation environment. Additionally, there is a need for an autonomous vehicle navigation or obstacle avoidance system to augment and / or prioritize navigation commands communicated to the vehicle. Summary of the Invention

[0005] Autonomous vehicles may be improved with flight control systems having multiple sensors (e.g., acoustic sensors, visual sensors, etc.) Multiple sensors may be employed in conjunction with collision avoidance solutions for small air vehicles, such as the "Panoptes™" collision avoidance system, more commonly referred to as dynamic obstacle avoidance systems.

[0006] According to a first aspect, a method for providing crash protection in an aircraft includes receiving a sensor input from a sensor positioned on the aircraft and operatively connected to a processor, the sensor configured to identify an obstacle within a field of view; receiving a pilot command stream from a pilot; identifying an obstacle within a field of view based at least in part on the sensor input; determining a region within a field of view in which the obstacle is located from a plurality of regions within the field of view, the region being determined at least in part based on the sensor input; setting a control input in response to the region determined in the determining step; inputting the control input to a proportional-derivative (PD) controller to generate control data; generating, via a processor, a control command stream in response to the control data; and comparing, via the processor, the control command stream to a pilot command stream to determine whether the pilot command stream from the pilot is safe.

[0007] According to a second aspect, a navigation system for providing crash protection in an aircraft comprises a sensor connected to the aircraft and configured to identify an obstacle within a visual field, a processor operatively connected to the sensor and a memory device, the processor configured to receive a pilot command stream from a pilot, and further configured to identify the obstacle within a visual field based at least in part on sensor input from the sensor, to determine an area in which the obstacle is located from a plurality of areas within the visual field, the area being determined at least in part on the sensor input, configured to set a control input in response to the area determined in the determining step, configured to input the control input to a proportional-derivative (PD) controller to generate control data, and configured, via the processor, to generate a control command stream in response to the control data, and configured, via the processor, to compare the control command stream with the pilot command stream to determine whether the pilot command stream from the pilot is safe or not.

[0008] In certain aspects, when the pilot command stream from the pilot is determined to be unsafe, a control command stream is communicated to a flight controller of the aircraft in place of the pilot command stream.

[0009] In certain aspects, when the pilot command stream from the pilot is determined to be safe, the pilot command stream is communicated to a flight controller for the aircraft.

[0010] In certain aspects, the sensor input includes a range-rate estimate or a range estimate.

[0011] In certain aspects, the pilot command stream is from a human pilot.

[0012] In a particular embodiment, the pilot command stream is from an autopilot.

[0013] In certain aspects, the plurality of regions includes a first region, a second region, and a third region, such as an incoming region, a panic region, and / or a critical region.

[0014] In certain embodiments, the first region is defined as the area between the sensor maximum distance and a first threshold (eg, an incoming threshold).

[0015] In certain embodiments, the second region is defined as the area between the first and second thresholds.

[0016] In certain embodiments, a third region is defined as the area between the second threshold (eg, the panic threshold) and the aircraft.

[0017] In certain aspects, the pilot command stream is determined to be unsafe if the pilot command stream can be interpreted by the processor as attempting to (1) reduce the distance between the aircraft and an obstacle or (2) increase the aircraft rate above an upper rate limit set by the control data.

[0018] In certain aspects, a pilot override command may be received from the pilot, which takes precedence over the control command stream.

[0019] In certain aspects, the processor is configured to perform a target filtering operation.

[0020] In certain aspects, the target filtering operations include receiving range and magnitude data for obstacles within the aircraft's line of sight from a RADAR system; determining, via a processor, whether a magnitude is met based at least in part on the range and magnitude data; calculating, via a processor, a standard deviation of at least a portion of a trace reflecting the range and magnitude data over time; determining, via a processor, a new range point for the trace; calculating, via a processor, a minimum difference between the new range point for the trace and a range assigned from the incoming data; and calculating, via a processor, a confidence and low pass value with a critically damped low pass filter (LPF).

[0021] In certain aspects, the confidence and low pass values ​​are calculated using a weighted average of statistical terms derived from the signal mean, standard deviation, and magnitude.

[0022] In certain aspects, the weighted average is defined by an operator relative to the desired filter performance.

[0023] According to a third aspect, a method for providing accuracy enhancing target filtering onboard an aircraft includes receiving range and magnitude data for obstacles within line of sight of the aircraft from a RADAR system; determining, via a processor, for each of a predetermined number of obstacles whether a magnitude is met based at least in part on the range and magnitude data, where if the magnitude is met the processor is configured to set the range to a known good value; calculating, via the processor, a standard deviation of at least a portion of a trace reflecting the range and magnitude data over time; determining, via the processor, a new range point for the trace; calculating, via the processor, a minimum difference between the new range point for the trace and a range assigned from the incoming data; determining, via the processor, whether each of a plurality of conditions is met, where if one or more of the plurality of conditions are not met the processor is configured to calculate a new filtered range point using linear regression; determining whether each of a plurality of conditions is met; incrementing an iteration counter; and calculating, via the processor, a confidence and low pass value with a critically damped low pass filter (LPF).

[0024] In certain aspects, each of the steps is performed or repeated for each of a predetermined number of obstacles within line of sight of the aircraft.

[0025] In a particular embodiment, the predetermined number of obstacles includes the five most prominent obstacles within line of sight of the aircraft.

[0026] In certain embodiments, the standard deviation is the standard deviation of the most recent 20 points of a trace through a 20-point linear regression.

[0027] In a particular embodiment, the minimum difference is the difference between the most recent distance of the trace and the assigned distance from the incoming data.

[0028] In certain embodiments, the conditions include (1) whether the smallest difference is greater than 3.5 times the standard deviation, and (2) whether the smallest difference is greater than 0.4.

[0029] In certain embodiments, the plurality of conditions further comprises: (1) whether the standard deviation is less than 0.2; and (2) whether the repeat counter is less than 15.

[0030] In certain aspects, the confidence and low pass values ​​are calculated using a weighted average of statistical terms derived from the signal mean, standard deviation, and magnitude.

[0031] In certain aspects, the weighted average is defined by an operator relative to the desired filter performance.

[0032] The above and other objects, features, and advantages of the devices, systems, and methods described herein will become apparent from the following description of specific embodiments thereof, in which the accompanying drawings are shown in which like numerals refer to like structure, and in which: The drawings are not necessarily drawn to scale, emphasis instead being placed upon illustrating the principles of the devices, systems, and methods described herein. [Brief description of the drawings]

[0033] [Figure 1] 1 illustrates an environment for autonomous navigation using a dynamic collision avoidance system. [Figure 2a] 1 illustrates a first exemplary autonomous vehicle having a dynamic collision avoidance system. [Figure 2b] 1 illustrates a second exemplary autonomous vehicle having a dynamic collision avoidance system. [Figure 2c] 1 shows an autonomous vehicle with its sensor positioned in a downward facing configuration. [Figure 2d]1 illustrates an autonomous vehicle with its sensors positioned in a forward-facing configuration. [Figure 2e] 1 illustrates an autonomous vehicle with its sensors positioned in both a forward-facing configuration and a downward-facing configuration. [Diagram 3] FIG. 1 is a block diagram of a dynamic collision avoidance and navigation system for an autonomous vehicle. [Figure 4] 1 is a flow chart of a method for using a dynamic collision avoidance system when navigating a vehicle from a location to a target. [Diagram 5] 5a-5c illustrate an exemplary RADAR Flight Control / Collision Avoidance (RFCA) module. [Figure 6a] 13 is a flow chart of an example three region collision protection function with pilot priority. [Figure 6b] A diagram of an aircraft facing three areas. [Figure 7] 4 is a flow chart of an exemplary landing assistance module. [Figure 8] FIG. 2 is a diagram of the input and output of an exemplary target filter. [Figure 9] 1 is a flow chart of an exemplary target filtering function. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] The embodiments of the present disclosure will be described below with reference to the accompanying drawings. In the following description, well-known functions or structures are not described in detail because they may obscure the present disclosure with unnecessary details. Described herein are devices, systems, and methods for autonomous vehicle navigation, in particular for navigation using multiple methods of obstacle avoidance.

[0035] All documents cited herein are incorporated herein by reference in their entirety. Unless expressly stated or clear from the context to the contrary, reference to a singular item should be understood to include the plural items, and vice versa. Unless stated otherwise or clear from the context, grammatical conjunctions are intended to express all alternative and conjunctive combinations of connected phrases, sentences, words, and the like. Thus, "or" should be generally understood to mean "and / or," and the like. The recitation of ranges of values ​​herein is not intended to limit the individual reference to any and all values ​​falling within the ranges, unless otherwise indicated herein, and each individual value within such range is incorporated herein as if it were individually recited herein.

[0036] Terms such as "about," "approximately," and the like, when attached to a numerical value, should be interpreted as indicating a deviation recognized by one of ordinary skill in the art to operate satisfactorily for the intended purpose. Values ​​and / or numerical ranges are provided herein as examples only and do not constitute limitations on the scope of the described embodiments. Any example or exemplary language used herein (such as, for example, or the like) is intended merely to better describe the examples and does not limit the scope of the embodiments. No language in the specification should be construed as indicating that any non-claimed element is essential to the implementation of each embodiment. In the following description, terms such as "first," "second," "top," "bottom," "side," "front," "rear," and the like, are used for convenience and should not be interpreted as limiting terms. With respect to this application, the following terms and definitions apply.

[0037] The terms "air vehicle" and "aircraft" refer to machines capable of flight, including, but not limited to, fixed-wing aircraft, unmanned aerial vehicles, variable-wing aircraft, and vertical take-off and landing (VTOL) aircraft.

[0038] As used herein, the terms "communicate" and "communicate" refer both to transmitting or otherwise conveying data from an origin to a destination, and to delivering data to a communications medium, system, channel, network, device, line, cable, fiber, circuit, and / or link for conveying the data to a destination.

[0039] The terms "circuits" and "circuitry" refer to physical electronic components (i.e., hardware) and any software and / or firmware ("code") that may configure, be executed by, and / or be associated with the hardware. As used herein, a particular processor and memory may comprise a first "circuit" when executing one or more lines of instructions of a first set, and a second "circuit" when executing one or more lines of instructions of a second set.

[0040] As used herein, the term "computer" refers to a programmable device designed to continuously and automatically perform sequences of mathematical or logical operations, including, but not limited to, personal computers (e.g., those available from Gateway®, Hewlett-Packard®, IBM®, Sony®, Toshiba®, Dell®, Apple®, Cisco®, Sun®, etc.), handhelds, processor-based devices, and any other electronic device equipped with a processor or microprocessor.

[0041] The term "exemplary" means serving as a non-limiting example, instance, or illustration. As used herein, the terms "eg" and "for example" highlight a list of one or more non-limiting examples, instances, or illustrations. As used herein, a circuit is "operable" to perform a function whenever the circuit comprises the necessary hardware and code (if necessary) to perform the function, regardless of whether the function is not possible or enabled to be performed (e.g., by operator-configurable settings, factory adjustments, etc.).

[0042] As used herein, the term "processor" refers to processing devices, apparatus, programs, circuits, components, systems, and subsystems, whether implemented in hardware, tangibly embodied software, or both, and whether or not programmable. As used herein, the term "processor" includes, but is not limited to, one or more computers, hardwired circuits, signal modifying devices and systems, devices and machines for controlling systems, central processing units, programmable devices and systems, field programmable gate arrays, application specific integrated circuits, systems on a chip, systems with discrete elements and / or circuits, state machines, virtual machines, and data processors.

[0043] As used herein, the term "navigation commands" refers to directives for guiding a vehicle. Navigation commands may be represented or provided as digital or analog data directives or signals by a flight control system. Navigation commands may be generated by, without limitation, an autopilot, a pilot (whether locally or remotely located), and / or an obstacle avoidance system. Navigation commands may be communicated to a controller or a flight control mechanism, for example.

[0044] The present disclosure seeks to provide systems and methods for facilitating autonomous vehicle navigation and / or obstacle avoidance by detecting collision threats. As disclosed herein, autonomous vehicle navigation and / or obstacle avoidance may be facilitated by detecting one or more collision threats using, among other things, acoustic techniques (e.g., echolocation sensors), visual techniques, or a combination thereof, to sense objects / obstacles (e.g., stationary and / or moving non-cooperative targets). Examples of such collision threats may include, but are not limited to, obstacles such as birds, humans, other vehicles, structures (e.g., buildings, bridges, towers, etc.), foliage (e.g., trees, grass, etc.), etc. While autonomous vehicle navigation and / or obstacle avoidance capabilities (e.g., any hardware and / or associated methodologies) may be incorporated into the air vehicle's control system (e.g., flight control system, whether human-piloted or autopilot-driven) during its initial design and fabrication, such capabilities may alternatively be provided via ancillary systems (e.g., add-on or "retrofit" systems) configured to control or override an existing flight control system. When ancillary systems are employed, they preferably do not require modifications to the existing flight control system (e.g., original navigation components) or air vehicle structure, thereby mitigating undesirable installation time and costs while continuing certification, permitting, etc. of any components.

[0045] In certain aspects, the autonomous vehicle navigation and / or obstacle avoidance functions may be vehicle agnostic and sensor agnostic. Indeed, in a typical small UAV, there is an overlap between the maximum range of the auditory and visual sensors and the closure rates for fixed obstacles or moving objects for which they are effective. Thus, the autonomous vehicle navigation, obstacle and / or collision avoidance functions may employ auditory and / or visual sensors for distance / range measurements. For example, UAVs and MAVs may be equipped with sensors, including electroacoustic, optical, RADAR, and / or those employing automatic dependent surveillance-broadcast ("ADS-B") (e.g., ADS-B receivers).

[0046] The autonomous vehicle navigation and / or obstacle avoidance system may be sensor agnostic and may fuse aggregated information (i.e., data) received from various sensors to generate a prediction of the global environment. Using the prediction of the global environment, features relevant to obstacle detection and navigation algorithms or collision avoidance algorithms may be extracted and stored in a database. The algorithm bank may access the database to determine if action must be taken to avoid a collision upon detecting a collision threat. The algorithm bank may also access the database to determine what action must be taken if action is deemed necessary.

[0047] If action is necessary, the autonomous vehicle navigation and / or obstacle avoidance system may interact with already existing vehicle infrastructure (e.g., existing flight control systems) to avoid collisions. Indeed, the interaction between the autonomous vehicle navigation and / or obstacle avoidance system and existing systems may be vehicle agnostic, thereby allowing the autonomous vehicle navigation and / or obstacle avoidance system to be connected to a variety of air vehicles, including already existing air vehicles.

[0048] FIG. 1 illustrates an exemplary environment 100 for autonomous navigation employing an obstacle avoidance system of the present disclosure. The system may also be generally referred to as a dynamic collision avoidance system and may facilitate an electronic bumper ("e-bumper") function. The environment 100 may include a target 102, one or more roads 110, and any number of obstacles, such as buildings 112, utility lines 114, utility poles 116, and trees 118. The environment 100 may further include unpredictable obstacles 122 along the path, which may be dynamically detected using the dynamic collision avoidance system. As shown, the air vehicle may be configured to follow one or more navigation paths (e.g., navigation paths 104, 106, 108) toward the target 102. Each path may be provided or determined, for example, via an autopilot, and may be configured to address one or more obstacles.

[0049] According to at least one embodiment, the airborne vehicle, whether guided by an autopilot or remotely controlled, may be configured to dynamically avoid unforeseen obstacles 122 using a dynamic collision avoidance system. In particular, upon detection of a collision threat (e.g., an unforeseen obstacle 122), the dynamic collision avoidance system may command the airborne vehicle, based on measurements received from multiple sensors, to override any commands from the autopilot or pilot (e.g., via a flight control system), avoid the unforeseen obstacle 122, and ultimately return to the navigation path.

[0050] 2a and 2b show perspective views of a vehicle (e.g., an autonomous vehicle) suitable for use with a dynamic collision avoidance system. The aircraft 200 shown in FIG. 2a may include an airframe 202, a landing gear 204, an electronics module 300 (shown in FIG. 3), and one or more thrust generators 206 (e.g., turbines, motors, or engines operably coupled to a propeller). The electronics module 300 may be integrated with the airframe 202 or provided via a separate housing or pod. FIG. 2b shows a second vehicle that is substantially the same as the aircraft 200 of FIG. 2a, but the sensor 210 of the second vehicle is positioned closer to the center of the vehicle and disposed in a separate housing 208. In particular, while one housing may be suitable for a particular application, two or more individual housings 208 (e.g., an attached navigation module as described in connection with FIGS. 5a-5c) may be positioned on the periphery of the aircraft 200 to provide a view oriented to the line of flight of the air vehicle. The individual housings 208 may be removable from the airframe 202 and may be further configured to house the electronic module 300, or portions of the electronic module 300 (e.g., function as a housing for the electronic module 300). Moreover, the functionality of the individual housings 208 may be provided in any suitable manner that does not require modifications to the original navigation components or structures resident on the air vehicle.

[0051] Thus, the sensors 210 may be positioned on the aircraft 200 in a downward facing configuration as shown in FIG. 2c to detect obstructions below, a forward facing configuration as shown in FIG. 2d to detect obstructions in front of the aircraft 200, or a combination thereof as shown in FIG. 2e. As can be appreciated, the sensors 210 may also be positioned on the sides, rear, and / or top of the aircraft 200 to detect obstacles and other threats in all directions relative to the aircraft 200. It should be noted, therefore, that the locations of the sensors 210 are determined by the designer as needed for a particular purpose, sensor type, and / or operation, and thus should not be limited to the layout depicted in this disclosure. The landing gear 204 may be simple skids as shown, or any other device, such as wheels, skids, skis, floats, or combinations thereof, that can support the aircraft 200 when it is not flying, while allowing the aircraft 200 to take off, land, and / or taxi without damage. The landing gear 204 may also be retractable to reduce drag during flight.

[0052] To facilitate controlled flight by adjusting the roll, pitch, and yaw of the aircraft 200, the aircraft 200 may further comprise one or more flight control mechanisms 304 or equivalent flight control systems configured to receive navigation commands and respond appropriately. To this end, the flight control mechanisms 304 may include one or more processors, actuators, motors, and / or other devices (e.g., electrical or electromechanical devices) operatively connected to a controller or capable of receiving and responding to navigation commands. Suitable flight control mechanisms 304 include, without limitation, conventional flight control surfaces (e.g., flaps, ailerons, elevators, rudders, spoilers, air brakes, and / or other flight control surfaces), as well as other flight control mechanisms, such as vectored-thrust control systems. Vectored-thrust control functions may be facilitated by moving the thrust generators 206 to direct thrust in a desired direction, thus controlling flight. For example, an articulated electric motor arrangement may employ vectored-thrust control to directly modify the thrust vector. Indeed, articulating the thrust vector motor pods independently enables rapid transitions between vertical and horizontal flight. In certain aspects, the aircraft 200 may further comprise two or more fins (e.g., vertical stabilizers and / or horizontal stabilizers), particularly for fixed-wing air vehicles.

[0053] Aircraft 200 may further comprise an intelligence, surveillance, and reconnaissance ("ISR") payload to gather data. For example, aircraft 200 may be equipped with a payload pod comprising one or more cameras, acoustic devices, and other sensors. Any video, imagery, acoustic, telemetry, and / or other sensor data ("survey data") collected by UAV 106 may be stored locally or wirelessly communicated in real time from aircraft 200 to a remote location using an antenna coupled with an on-board wireless communication device, such as a transceiver. Alternatively, the survey data may be communicated or otherwise conveyed (e.g., after operation, while tethered or on the ground) via a wired connection to a remote location or another party.

[0054] Although the vehicle 200 depicted in Figures 2a-2e is a vertical take-off and landing ("VTOL") air vehicle, it should be understood that the autonomous vehicles described herein may include any vehicle, device, component, element, etc. that can be usefully navigated using the principles of the dynamic collision avoidance system disclosed herein, including, without limitation, any unmanned vehicle, manned vehicle, air vehicle, ground vehicle, surface water vehicle, space vehicle, remotely controlled vehicle, large vehicle, small vehicle, etc., unless otherwise specified or apparent from the text. For example, the autonomous vehicles described herein may include helicopters or other vehicles that use horizontal propellers for lift, etc. The autonomous vehicles described herein may additionally or alternatively include air vehicles with forward flight capabilities, such as fixed-wing air vehicles. For further information, other suitable autonomous vehicles are disclosed in more detail in commonly owned U.S. Patent No. 8,500,067, entitled "Modular Miniature Unmanned Aircraft With Vectored-Thrust Control," and U.S. Patent Publication No. 2015 / 0260526, entitled "Autonomous Vehicle Navigation System And Method." For example, U.S. Patent Publication No. 2015 / 0260526 describes an aircraft and sensor payload that provides an improved navigation system that benefits from both cameras and echolocation sensors with overlapping fields of view.

[0055] In general, the electronics module may be used to house the vehicle's avionics, power sources (e.g., propulsion batteries, generators, etc.), sensor payload, and communication devices or systems. As described above, the electronics module may be integrated with the airframe 202 or may be contained within a separate housing. The separate housing may also provide rigidity to the airframe 202. Thus, the electronics module may be removable from and replaceable with the airframe 202 and may house any system or subsystem of the electronic bumper and / or navigation system as contemplated herein. The electronics module may include electronics and hardware used to support or facilitate the electronic bumper and navigation systems and methods. However, certain electronics and / or hardware may be configured outside of the electronics module housing. For example, the aircraft 200 may further include one or more sensors 210 used to facilitate autonomous flight. The one or more sensors 210 may include, without limitation, echolocation sensors, ultrasonic sensors, infrared sensors, RADAR, etc. Sensors 210 may be appropriately installed on aircraft 200 to enable functionality. For example, placement of certain sensors (e.g., visual or acoustic-based sensors) may be configured on aircraft 200 outside of the electronics module housing (if used) because placing the certain sensor within the electronics module housing would impede or interfere with sensor function. For example, as shown in Figures 2a and 2b, sensors 210 may be positioned on a surface (e.g., top, bottom, edges, etc.) of airframe 202 and / or on the electronics module housing (e.g., individual housing 208).

[0056] Sensors 210 may employ one or more echolocation sensors. These sensors generally function by emitting audio frequencies into the environment and detecting any echoes of the audio frequencies that return from an obstacle to the vicinity of the echolocation sensor. Using the strength of the echoes and / or the direction of the echo returns, the echoes may be used to find and / or identify the obstacles. The (identified) obstacles may in turn cause the airborne vehicle to change direction to avoid colliding with the one or more obstacles.

[0057] Regardless of the type of sensor 210 employed, the dynamic collision avoidance system may be configured to override or dampen commands from a remote pilot when such commands could cause the aircraft 200 to collide with an obstacle. Thus, the dynamic collision avoidance system provides (1) damping of operator inputs that could lead to a collision, and (2) active reduction of the velocity component in the direction of the object, if necessary.

[0058] To this end, sensors may be positioned to obtain a view in the direction of vehicle travel, thereby identifying potential obstacles in the path of aircraft 200. For example, a single sensor (or a single group of sensors) may be provided at the front of the vehicle to detect collision threats (e.g., obstructions or obstacles) in the path of the vehicle. Additionally, multiple sensors 210 (or multiple groups of sensors) may be positioned around the perimeter (and / or above and below) of aircraft 200 to provide a view oriented toward the line of flight of aircraft 200. Thus, multiple sensors 210 may enable aircraft 200 to detect collision threats on any side of aircraft 200.

[0059] As described herein, the sensor 210 may include any vision-based or echolocation sensor known or to become known in the art, including, but not limited to, ultrasonic sensors, among others. In one aspect, the camera 206 may be used to identify larger objects through three-dimensional reconstruction techniques such as optical flow. While this may provide useful information for autonomous navigation, the processing latency associated with optical imaging, as well as sensitivity to the visibility of various types of objects, may limit the utility of optical sensing techniques to detect small, rapidly approaching objects in the vehicle's line of flight. By directing the sensor 210 toward the line of flight, acoustic detection may be used to complement optical detection and detect emergent obstructions that should trigger the vehicle to perform responsive maneuvers.

[0060] It will be appreciated that one purpose of the acoustic sensor is to provide instantaneous detection of obstacles that are directly within the flight path (or other line of travel), particularly obstacles that may not be detected using visual detection or other techniques. Accordingly, it will be appreciated that one purpose of the sensor 210 is to provide instantaneous detection of obstacles in a particular direction (e.g., in any direction of the vehicle), particularly obstacles that may not be reliably detected using visual detection or other techniques. While echolocation arrays work well in this context, additionally or alternatively, other sensor systems, such as laser-based technology or any other suitable technology using optical, acoustic, radio frequency, or other sensing modalities, may also be suitably employed for rapid and accurate detection of obstacles. Any such technology suitable for implementation within an autonomous vehicle may accurately and quickly identify obstructions and may be used in place of echolocation sensors in the systems and methods contemplated herein. Thus, the dynamic collision avoidance system is generally sensor agnostic and may be configured to employ one of a variety of sensor technologies or combinations thereof in that regard. For example, the dynamic collision avoidance system may employ a combination of vision-based and acoustic-based sensors.

[0061] The electronics module may be provided as a single housing, but instead, the electronics module may comprise multiple housings or "sub-housings." For example, the electronics module may be split into two housings: a first housing for heavier components such as batteries, and a second housing for more delicate components such as avionics, survey payloads, sensor payloads, and any other electronic equipment. The components may be distributed or split between the housings to provide a desired weight distribution across the airframe 202.

[0062] The flight control system may be used to control and / or navigate the aircraft 200. The flight control system need not be a separate physical item on the vehicle, but rather may be a component of a larger navigation system or may itself include the entirety of navigation system components. Unless otherwise specified or clear from the context, any components described with reference to a navigation system may also be used by or included within a flight control system, and vice versa. In operation, the flight control system determines and / or commands the aircraft 200 to follow a navigation path to reach a desired location based on signals received from the navigation system components. For example, the flight control system may facilitate autopilot function and / or respond to remote navigation commands. To this end, the flight control system 306 may be configured to communicatively connect the aircraft 200 to a remote location and to send and receive (e.g., send and receive) signals between the aircraft 200 and the remote location. The functionality of the navigation module may be distributed in any suitable manner among components within the flight control system, other components within aircraft 200, and / or remotely located components. Additionally, suitable electronic, mechanical, and communication interfaces may be provided to facilitate removal and replacement of the electronic module from airframe 202.

[0063] FIG. 3 is a block diagram of an aircraft 200 (e.g., an autonomous vehicle) having a flight control system 306, a dynamic collision avoidance system 302, an electronics module 300, and a flight control mechanism 304. More specifically, FIG. 3 illustrates an electronics module 300 that is used to house or otherwise contain the vehicle's flight control system 306, a power source 336 (e.g., a propulsion battery), a sensor payload (e.g., an ISR payload 334), and a communication device(s) 338. However, even though a particular arrangement is shown in FIG. 3, it will be understood that the arrangement of components may vary. For example, the flight control system 306 and / or the dynamic collision avoidance system 302 may be located in one or more dedicated housings and / or may be removable from the aircraft 200. For example, the functionality of the dynamic collision avoidance system may be provided via an integrated navigation module that is removably and non-permanently connected to the vehicle (e.g., via the airframe). Such an integrated navigation module may be configured to intercept and modify signals or navigation commands as disclosed herein.

[0064] Alternatively, flight control system 306 and / or dynamic collision avoidance system 302 may be integrated within aircraft 200 and communicatively coupled to electronics module 300 and / or flight control mechanism 304. In certain embodiments, flight control system 306 and / or dynamic collision avoidance system 302 may share components, such as memory, sensors, processors, or controllers. Additionally, electronics module 300 may be removably coupled to aircraft 200 or integrated into the fuselage of aircraft 200, or the like, in any desired manner. Thus, the placement of the various components may be configured as desired by a designer or operator, and thus should not be limited to the particular examples described or illustrated herein. For example, flight control system 306 and / or dynamic collision avoidance system 302 may be mounted to the exterior of the vehicle or located entirely or partially within the confines of the vehicle. The flight control system 306 and / or the dynamic collision avoidance system 302 may be a removable and replaceable package or a module that is removable from and replaceable to the vehicle, or may be permanently connected to or integrated into the vehicle.

[0065] The modular housing may encase one or more components of electronics module 300, flight control system 306, and / or dynamic collision avoidance system 302. The modular housing may be constructed from plastic, metal, wood, composites, ceramics, or any material appropriate for a particular vehicle purpose or type of vehicle. The modular housing may be removable or removable, or may be permanently connected to the vehicle. The modular housing may be attached to the vehicle in any manner known to one of ordinary skill in the art. The modular housing may include openings for sensors, such as sensor 210.

[0066] Electronic Module 300 As mentioned above, electronics module 300 is used to house the vehicle 200's avionics (e.g., flight control system 206), power source 336, sensor payloads such as ISR payload 334, and communications devices or systems 338, and may be integrated with airframe 202 or contained within a separate housing. In certain aspects, electronics module 300 may further comprise a dynamic collision avoidance system 302, or functionality of a dynamic collision avoidance system 302.

[0067] Steering mechanism 304 Steering mechanism 304 may be configured to steer aircraft 200 (whether autonomous or human controlled) on a navigation path to a destination as contemplated herein. Aircraft 200 may be any vehicle referenced herein or otherwise known in the art (or that becomes known in the art). Similarly, steering mechanism 304 may be in the form of any steering referenced herein or otherwise known in the art (or that becomes known in the art). Generally, steering mechanism 304 responds to signals from flight steering system 306. Flight steering system 306 may employ feedback or other control systems to precisely guide aircraft 200 along an intended path.

[0068] As discussed above, the flight control mechanism 304 may include, for example, a rudder and elevators at the rear of the aircraft 200, and may also include any other suitable control surfaces for a vertically flying vehicle. Such control surfaces may involve coupled cables, actuators, and the like. The flight control mechanism 304 may additionally or alternatively include any mechanism for piloting an autonomous vehicle. For example, for an air vehicle, the flight control mechanism 304 may more generally include rudder, elevators, flaps, ailerons, spoilers, air brakes, and other control surfaces. For other air vehicles, such as helicopters, the flight control mechanism 304 may include several rotors along with foils and other control surfaces. Some rotors may be fixed rotors or movable rotors. The flight control mechanism 304 may also include articulated electric motors employing vectored thrust control to directly modify the thrust vector. For ground-based vehicles, the flight control mechanism 304 may include rack and pinion systems, variable rotation treads, recirculating ball systems, and the like. Steering mechanism 304 may additionally or alternatively include any components that provide thrust, acceleration, and deceleration of air vehicle 200, along with directional control. Although the vehicle may typically use separate or integrated components for drive and guidance, all such combinations that facilitate control over the vehicle's motion are intended to fall within the scope of "steering mechanism" as contemplated herein.

[0069] Dynamic Collision Avoidance System 302 The electronic bumper module generally includes circuitry to facilitate the electronic bumper function of the obstacle avoidance system. In fact, the flight control system 306 and the dynamic collision avoidance system 302 may cooperate to provide an obstacle avoidance system. As disclosed herein, the dynamic collision avoidance system 302 may include one or more sensors 210, where each sensor 210 may have an acoustic field of view ("FOV"). However, the sensors 210 may be any of the echolocation sensors referred to herein or otherwise. Using these sensors 210, the dynamic collision avoidance system 302 may detect unpredictable obstacles 122 and communicate responsive navigation commands to the flight control system 306 to avoid the unpredictable obstacles 122.

[0070] For example, as shown in Figures 2a and 2b, the functionality of the dynamic collision avoidance system 302 may be facilitated via the flight control system 306 or a separate system using a processor 340 (or other comparable logic), memory 342, and one or more sensors 210 (e.g., acoustic sensors, visual sensors, or a combination thereof) positioned along the top, bottom, and / or perimeter (e.g., one or more edges) of the body 202 of the aircraft 200. The dynamic collision avoidance system 302 may be used to reduce the likelihood of a collision with an obstacle in any orientation of the aircraft 200 and for any relative position of the object to the vehicle. More specifically, the dynamic collision avoidance system 302 may be provided through multiple sensors 210. The multiple sensors 210 may be used to detect various obstacles. In general, the dynamic collision avoidance system 302 may communicate directly with the flight control mechanism 304 (or via a controller) and / or the flight control system 306, for example, to provide sensed data from the sensors 210 and / or derived commands (e.g., modified navigation commands, such as alternate navigation paths, attenuated navigation signals, or responsive maneuvers, which may be control commands responsive to the sensed data or predictions of the global environment and configured to avoid unpredictable obstacles 122). Thus, the dynamic collision avoidance system 302 is particularly useful in missions that include operation near obstacles.

[0071] Specific benefits and contributions of the dynamic collision avoidance system 302 include: (1) a sensor-agnostic methodology that may be used to generate predictions of the global environment based on individual sensor inputs, (2) a sensor-agnostic methodology that may be used to interact with existing vehicle control infrastructure in a vehicle-agnostic approach, and (3) navigation algorithms necessary to perform electronic bumper functions. For example, the dynamic collision avoidance system 302 may be integrated into the aircraft 200 and communicatively coupled to the flight control mechanisms 304, the flight control system 306, the optical system, the sensors 210, or a combination thereof.

[0072] The dynamic collision avoidance system 302 is also advantageous in that it employs a relatively simple state machine that activates, scales, or deactivates the influence of each component, i.e., operator input, proportional integral derivative ("PID") over distance, autopilot command, etc., depending on the distance from the object, thereby reducing the risk of error. Furthermore, the prediction of the complete state machine can be compiled using as few as four echolocation sensors. However, in certain embodiments (e.g., when only one direction needs to be monitored), an obstacle avoidance system can be provided using only a single sensor placed at the front end of the vehicle. Another advantage of the dynamic collision avoidance system 302 as disclosed herein is that the dynamic collision avoidance system 302 does not require any cooperating target sensors. That is, there is no need to place a corresponding sensor on the obstacle, thereby greatly enhancing the usefulness of the dynamic collision avoidance system 302. Furthermore, the dynamic collision avoidance system 302 does not require air vehicle data information or collision avoidance algorithms.

[0073] Various physical configurations are possible, and the dynamic collision avoidance system 302 may additionally or alternatively be integrated into the vehicle 300, the flight control system 306, or include any of the components described herein. To this end, as described with respect to FIGS. 2a and 2b, the sensor 210 may be integrated within the hull of the vehicle 300. The integration of the sensor 210 provides several advantages. For example, the integration of the sensor 210 may provide a compact package (e.g., in size and weight) while avoiding interference (crosstalk) of echolocation sensors, and / or electromagnetic interference ("EMI") and propeller acoustic noise. Furthermore, the hull of the aircraft 200 allows for precise placement, low drag forces, and easy replacement of the sensor 210 (if the sensor is damaged or it is otherwise desirable to replace / update the sensor). For example, recesses may be provided within the hull of the vehicle for one or more sensors, thereby reducing undesirable drag forces. The sensor may further be covered with a protective cover. However, the protective cover should be constructed so as not to interfere with the function / reliability of the sensor. For example, when an acoustic-based sensor is employed, the protective cover should be acoustically invisible (e.g., a fabric or reflector with small holes at regular intervals covered by a thin elastic membrane). An example of an acoustically invisible reflector material is described in Phys.Rev.Lett.110, 244302 (June 13, 2013) by Jong Jin Park et al. entitled "Giant Acoustic Concentration by Extraordinary Transmission in Zero Mass Metamaterials". Similarly, when a visual-based sensor is employed, the protective cover should be transparent or otherwise designed to be visible.

[0074] While the current dynamic collision avoidance system 302 is described as using echolocation sensors as sensors 210, the dynamic collision avoidance system 302 can take measurements received from any sensor (whether, without limitation, echolocation sensors as described herein or of another type) and fuse the received data to generate a prediction of the global environment. From the prediction of the global environment, features of interest to the algorithm can be extracted and stored in a target database. The prediction of the global environment can be a summary of what the sensors are detecting around the aircraft. For example, if multiple sensors are available that provide measurements of the same state (e.g., distance to an obstacle), the state values ​​are fused. If multiple sensors are available that provide multiple different states (e.g., distance to an obstacle and speed / occlusion rate of the obstacle). As described above, by illustrating this prediction of the global environment as a summary of sensed data, the target database acts as a single interface accessible by a set of algorithms.

[0075] This target database may act as a common interface to any algorithms that may be used by the dynamic collision avoidance system 302 or another autonomous vehicle navigation or obstacle avoidance system. In the case of the dynamic collision avoidance system 302, objects determined to pose a collision threat are passed to the dynamic collision avoidance system's 302 dedicated algorithms.

[0076] As described above, the dynamic collision avoidance system 302 may be integrated into the aircraft 200 via a vehicle control interface of the flight control system 306 without modifying existing systems of the aircraft 200 and without requiring knowledge of the state of the vehicle's autopilot (or human control). For example, an embedded logic device or processor 340, such as an Arduino© microprocessor, may (1) intercept original navigation commands (e.g., flight commands by the pilot or autopilot), (2) dampen them according to a predefined electronic bumper algorithm, and (3) provide new or modified navigation commands (e.g., dampened commands generated by the dynamic collision avoidance system) to the vehicle's flight control system 306 (e.g., autopilot) or flight control mechanism 304 as replacement or preferred navigation commands. Additionally, the dynamic collision avoidance system 302 intercepts control signals (e.g., navigation commands) from the autopilot to the thrust generator 206 (e.g., an electric motor via a pulse width modulated ("PWM") signal) and modifies those signals before sending those modified signals to the flight control system 306 and / or the thrust generator 206 (e.g., a motor, turbine, etc.). A benefit of the presently disclosed dynamic collision avoidance system 302 is that it achieves the objective of obstacle avoidance without requiring modifications to the aircraft 200, i.e., the dynamic collision avoidance system 302 is vehicle agnostic. In certain aspects, software implemented in the embedded device of the dynamic collision avoidance system 302 may be constantly monitored for execution speed, since significant delays in the system may result in undesirable operation of the aircraft 200.

[0077] With global environment predictions and vehicle control interfaces in place and adjusted, the dynamic collision avoidance system 302 can focus on navigation commands (such as those from the pilot or autopilot) that may cause the vehicle to collide with the object. A PID distance controller can also hold the (vehicle) position and reject undesirable operator inputs (e.g., obstacle avoidance). Pilot inputs (e.g., navigation commands) can be rejected or modified using high-pass, low-pass, and band-pass filters, feed-forward approaches, and high-gain integrators. Additionally, gain scheduling techniques are implemented for robust control. For example, if the vehicle approaches an unpredictable obstacle 122 due to control inputs by the vehicle operator, the dynamic collision avoidance system can reduce the effectiveness of those control inputs depending on the distance to the object. If the vehicle continues to approach the unpredictable obstacle 122, the dynamic collision avoidance system will eventually fully damp all control inputs in the direction of the object and / or actively reduce the speed at which the object is approaching. Similarly, if environmental conditions (e.g., wind or gusts) cause the vehicle to move towards an object creating a risk of collision, the dynamic collision avoidance system provides navigation commands to ensure that the vehicle's position does not come within less than a predetermined safe distance from the object. Damping may be adjusted so that the dynamic collision avoidance system reduces the effectiveness of their control inputs (e.g., navigation commands) depending on the distance to the object. For example, an inverse distance-damping function may be employed, whereby as the distance between the vehicle and the object decreases, the control input damping increases such that the control input is effectively reduced, ignored, or reversed. Furthermore, in certain conditions, the dynamic collision avoidance system may be engaged to maintain the vehicle at a constant distance from the object to allow the object to be systematically surveyed at close range without the risk of impacting an unpredictable obstacle 122. The dual mode capabilities of the dynamic collision avoidance system 302, including both the ability to maintain distance to an object or the ability to perform obstacle avoidance in flight, make the dynamic collision avoidance system 302 useful to a wide range of operators.That is, maintaining distance between the vehicle and the object is advantageous for data collection, while obstacle avoidance assists inexperienced pilots.

[0078] The dynamic collision avoidance system 302 may further include an arming / disarming feature. The arming / disarming feature may be used, for example, to prevent the electronic bumper from falsely signaling a collision when it detects the ground during takeoff and landing. Indeed, a safety measure of arming and disarming the dynamic collision avoidance system 302 during various phases of flight, takeoff, and landing further enhances the robustness and safety of the aircraft 200. For example, the dynamic collision avoidance system 302 may be manually actuated by a controller (e.g., a pilot operates the vehicle) or may be automatically armed / disarmed depending on the vehicle's distance, location, altitude, flight time, etc. In certain aspects, the dynamic collision avoidance system 302 may be configured to be switched between one of a number of operating modes. The desired operating mode may be selected, for example, using a physical switch positioned on the air vehicle, remotely via an operator interface / device (e.g., via a remote controller / ground point), or the like. Exemplary operational modes include, without limitation, a non-operational mode (i.e., the system is turned off), a precision mode, a performance mode, and the like. For example, in precision mode, the dynamic collision avoidance system 302 may enable features such as automatic takeoff, rejection of erroneous operator inputs, obstacle avoidance, and precise control of the aircraft through adjusted input control / aircraft response mapping. Precision mode may also be designed to allow the UAV to approach obstacles closer (e.g., about 1 to 10 feet, more preferably about 3 to 7 feet, or about 4.5 feet) than in performance mode. In performance mode, the dynamic collision avoidance system 302 may provide the same benefits as precision mode, but may be optimized for faster flight regimes. For example, performance mode allows the aircraft to avoid objects at greater distances than precision mode. Additionally, avoidance maneuvers may be more aggressive than precision mode maneuvers to protect against faster operating speeds.

[0079] Flight Control System 306 Flight control system 306 may determine one or more navigation paths for aircraft 200 to reach a desired location based on signals received from navigation system components. Flight control system 306 may calculate, generate, and send navigation commands (e.g., data signals) to flight control mechanisms 304 to guide aircraft 200 along the navigation paths to a desired location. Flight control system 306 may be located, in whole or in part, within a separate housing, within airframe 202, or a combination thereof. Flight control system 306 may further include, for example, any of the components of dynamic collision avoidance system 302 or flight control system 306 described with reference to FIG. 3 . Indeed, flight control system 306 and dynamic collision avoidance system 302 are generally configured to direct or otherwise control one or more flight control mechanisms 304 in aircraft 200. Flight control system 306 may be communicatively coupled to aircraft 200 and the remote locations and configured to transmit signals to and receive signals from aircraft 200 and the remote locations via communications device 338. For example, communications device 338 may be a wireless transceiver and antenna.

[0080] In general, flight control system 306 may include a piloting system 308, a map system 310, a GPS system 312, a processor 314, a gyroscope 316, a controller 318, an accelerometer 320, and / or a memory 330. Flight control system 306 may also include the components described above as being disposed within the housing of electronics module 300, such as any other conventional flight instruments, sensors, processing circuitry, communication circuitry, optical systems including cameras, etc., necessary or useful for operation of an unmanned aerial vehicle or other autonomously or manually piloted vehicle, as well as other sensors 332. One or more of the components of flight control system 306 may be housed within the housing of electronics module 300.

[0081] Flight control system 306 may be communicatively coupled to one or more flight control mechanisms 304 and / or dynamic collision avoidance system 302. For example, flight control system 308 may be configured to receive signals from flight control system 306 (or dynamic collision avoidance system 302) and provide appropriate control signals to vehicle flight control mechanisms 304 in order to guide aircraft 200 along an intended path.

[0082] The map system 310 can be part of a map-based flight control system that provides location information about natural and man-made features in an area. This can include information at any level of detail, including, for example, a topographical map, i.e., a general two-dimensional map identifying roads, buildings, rivers, etc., or detailed three-dimensional data characterizing the height and shape of various natural and man-made obstructions, such as trees, statues, public infrastructure, buildings, etc. In one embodiment, the map system 310 can cooperate with an optical system for visual verification of the surroundings, or the map system 310 can cooperate with a GPS system 312 to provide information about various obstacles in the environment for purposes such as route determination. In one embodiment, the map system 310 can provide supplemental navigational assistance in a GPS-denied or GPS-impaired environment. When GPS is partially or completely unavailable, the map system 310 can cooperate with other sensors 332, such as optical sensors, inertial sensors, etc., to provide location information until a GPS signal can be restored.

[0083] Map system 310 may more generally communicate with other components of flight control system 306 to support navigation of the vehicle contemplated herein. This may include providing map information for route calculation, but may also include independent navigation capabilities. For example, map system 310 may provide a map-based navigation system that stores a map of an operating environment including one or more objects. The map-based navigation system may be connected to a camera and configured to determine the location of the vehicle by comparing stored objects to the visible environment. The visible environment may provide location data in the absence of GPS data or other location information.

[0084] GPS system 312 may be part of a global positioning system configured to determine a position of electronics module 300 or aircraft 200. GPS system 312 may include any GPS technology known in the art or that becomes known in the art, including conventional systems, satellite-based systems, and other systems that use publicly or privately operated beacons, location signals, and the like. GPS system 312 may include one or more transceivers that detect data for use in calculating a position. GPS system 312 may cooperate with other components of flight control system 306 to control the operation of aircraft 200 and navigate the vehicle along an intended path.

[0085] Gyroscope 316 may be a device configured to detect rotation of electronics module 300 or aircraft 200 to which electronics module 300 is connected. Gyroscope 316 may be integrated with aircraft 200 or may be located inside or outside the housing of electronics module 300. Gyroscope 316 may include any gyroscope or variations thereof known or to become known in the art (e.g., gyrostats, microelectromechanical systems (“MEMS”), fiber optic gyroscopes, vibratory gyroscopes, dynamically tuned gyroscopes, etc.). Gyroscope 316 may cooperate with other components of flight control system 306 to control the operation of aircraft 200 and navigate the vehicle along an intended path.

[0086] Accelerometer 320 may be any device configured to detect linear motion of electronics module 300 or aircraft 200. Accelerometer 320 may be integrated with aircraft 200 or may be located inside or outside the housing of electronics module 300. Accelerometer 320 may include any accelerometer known in the art (e.g., capacitive, resistive, spring mass-based, direct current ("DC") response, electromechanical servo, laser, magnetic induction, piezoelectric, optical, low frequency, pendulum type integrating gyro accelerometer, resonant, strain gauge, surface acoustic wave, MEMS, thermal, vacuum diode, etc.) or any accelerometer that becomes known in the art. Accelerometer 320 may cooperate with other components of flight control system 306 to control the motion of aircraft 200 and navigate the vehicle along an intended path.

[0087] Other sensors (or sensor systems) 332 or sensors 210 may be employed as well. For example, aircraft 200 (or the vehicle's flight control system 306, dynamic collision avoidance system 302, or electronics module 300) may employ infrared sensors, RADAR (i.e., radio detection and ranging) sensors, LiDAR (i.e., light detection and ranging) sensors, etc. Any of the above may be used alone or in combination with other systems and sensors described herein to augment the navigation of the vehicle. Processor 314 may be communicatively coupled to controller 318, aircraft 200, flight control system 306, flight control mechanism 304, and various other components, systems, and subsystems described herein. Processor 314 may be an internal processor of aircraft 200 or flight control system 306, an additional processor in electronics module 300 supporting various navigation functions contemplated herein, a processor such as a desktop computer connected locally or remotely to aircraft 200 and flight control system 306, a server or other processor connected to aircraft 200 and flight control system 306 via a data network, or any other processor or processing circuitry. In general, processor 314 may be configured to perform various processing and computational functions to control operation of aircraft 200 or flight control system 306 and support navigation. Processor 314 may include several different processors that cooperate to perform the steps described herein, such as when an internal processor of aircraft 200 controls operation of aircraft 200 while a processor within the housing pre-processes light data and echolocation data.

[0088] The processor 314 may be configured to determine or revise a navigation path for the aircraft 200 to a location based on various inputs including, for example, position information, movement information, dynamic collision avoidance system 302 data, etc. Those inputs may be variously based on the GPS system 312, the map system 310, the gyroscope 316, the accelerometer 320, and any other navigation inputs, as well as the optical system and the echolocation system. The optical system and the echolocation system may provide information about obstacles in the environment around the aircraft 200. For example, an initial path may be determined based solely on position information provided by the GPS system 312, with in-flight adjustments based on movements detected by the gyroscope 316, the accelerometer 320, etc. The processor 314 may also be configured to utilize an optical navigation system. In that case, the processor is configured to process a sequence of images using, for example, optical flow and identify visible obstacles within the FOV of the optical system to avoid the GPS system 312 and navigate the aircraft 200 to a location around the visible obstacles. Processor 314 may be further configured to identify obstacles within the FOV of dynamic collision avoidance system 302, typically within the vehicle's line of flight, and perform responsive maneuvers to guide aircraft 200 around the obstacles, avoiding the GPS system 312 and optical navigation system, and return aircraft 200 to a previous course toward a position.

[0089] The controller 318 may be operable to control components of the aircraft 200 and flight control system 306, such as the flight control mechanisms 304. The controller 318 may be electrically or otherwise communicatively coupled to the processor 314, the aircraft 200, the flight control system 306, the flight control mechanisms 304, and various other components of the devices and systems described herein. The controller 318 may include any combination of suitable software and / or processing circuitry to control the various components of the aircraft 200 and flight control system 306 described herein, including, without limitation, microprocessors, microcontrollers, application specific integrated circuits, programmable gate arrays, and any other digital and / or analog components, as well as combinations of the foregoing, with inputs and outputs for communicating control signals, drive signals, power signals, sensor signals, and the like. In one aspect, it may include circuitry directly and physically coupled to the aircraft 200 and flight control system 306, such as an on-board processor. In another aspect, it may be a processor, such as the processor 314 described herein. The processor may be coupled, for example, to a personal computer or other computing device connected to the aircraft 200 and flight control system 306 through a wired or wireless connection. Similarly, various functions described herein may be allocated between on-board processors for the aircraft 200, the flight control system 306, and separate computers. All such computing devices and environments are intended to be included within the meaning of the term "controller" or "processor" as used herein, unless a different meaning is expressly stated or is otherwise clear from the context.

[0090] Memory 330 may include local memory or remote storage devices that store logs of data for flight control system 306, including, but not limited to, positions of detected obstacles, maps, images, orientation, speed, navigation path, maneuver specifications, GPS coordinates, sensor readings, and the like. Memory 322 may additionally or alternatively store logs of data collected from several navigations of a particular vehicle, or several navigations of different vehicles. Memory 322 may additionally or alternatively store sensor data from optical system and dynamic collision avoidance system 302, associated metadata, and the like. Data stored in memory 330 may be accessed by processor 314, controller 318, a remote processing resource, and the like.

[0091] 4 is a flow chart of a method 400 for navigating a vehicle using a dynamic collision avoidance system. The dynamic collision avoidance system begins or is activated at step 402. Once activated, the dynamic collision avoidance system monitors the environment (e.g., in each direction of the degrees of freedom) using one or more sensors. As mentioned above, the sensors may be, for example, echolocation sensors. As shown at step 404, the method 400 may include detecting an unforeseen obstacle 122.

[0092] As shown at step 404, method 400 may include detecting an obstacle using one or more echolocation sensors (or other suitable sensors). Step 404 may include outputting acoustic signals, detecting echoes of those acoustic signals, and using the detected echoes to determine the size and location of the obstacle. Generally, this may be any obstacle that can be detected through an acoustic stream that blocks, partially blocks, obscures, compromises, etc. the vehicle's navigation path from its location to the target. The obstacle may be any physical obstruction, such as a building, tree, power line, rock, etc. More generally, the first obstacle may be any location or path that the vehicle should avoid.

[0093] As shown in step 406, method 400 may include determining whether to dampen the navigation command. The decision as to whether to dampen the navigation command in step 412 or calculate a responsive maneuver in step 408 may be based on the distance of the obstacle. For example, if the distance to the obstacle meets or exceeds a predetermined distance threshold, the dynamic collision avoidance system processor may dampen the navigation command in step 412. If the distance to the obstacle is less than the predetermined distance threshold, indicating a more imminent collision, the dynamic collision avoidance system processor may calculate a responsive maneuver in step 408.

[0094] As shown at step 412, the method 400 may include damping the navigation command. For example, if the vehicle approaches an obstacle due to a control input by the vehicle operator or the autopilot, the dynamic collision avoidance system may adjust the control input based on the distance to the object. If the vehicle continues to approach the obstacle, the dynamic collision avoidance system may eventually completely reject all control input in the direction of the object and / or actively reduce the speed at which the object is approaching.

[0095] As shown at step 414, the method 400 may include calculating a reactive maneuver to avoid the obstacle. In one aspect, the reactive maneuver may be a predefined reactive maneuver that provides a temporary deviation from the revised course and immediately returns to the revised course after the reactive maneuver is executed. In another aspect, this may include selecting among several predefined reactive maneuvers according to information about the obstacle, or dynamically generating a reactive maneuver according to feedback from the echolocation system. If necessary, the reactive maneuver may be further adapted to other data, such as GPS data, optical data, or other sensor data, to better react to the detected obstacle situation. Whatever calculation is performed, an instruction to calculate the reactive maneuver may be sent to the vehicle's steering system for corresponding execution.

[0096] As shown at step 410, method 400 may include determining whether the obstacle is out of range as a result of damping at step 412 or reactive maneuver at step 414. If the obstacle is out of range, method 400 may end at step 416, or alternatively may restart at step 402 to avoid future obstacles. If the obstacle is still within range, method 400 may return to step 404.

[0097] RADAR Flight Control / Collision Avoidance (RFCA) Module 500 The dynamic collision avoidance system 302 may be embodied as a RADAR Flight Control / Collision Avoidance (RFCA) module 500 that uses a RADAR as one of the sensors 210, such as a micro RADAR sensor with a rectangular beamformer. As shown in Figures 5a-5c, the RFCA module 500 may include a radio frequency large scale integration (RFLSI) component 502, one or more memory devices 504 (e.g., RAM and ROM) and a microcontroller (MCU) 508 connected to one or more of the sensors 210, a cable input 506, such as an on-the-go USB connector, and an interface connector 510 with the aircraft 200. The RFCA module 500 may further include various electronic components 512 (e.g., capacitors, coils, low dropout (LDO) regulators, etc.) for handling signals and voltages. One of the sensors 210 (eg, a micro RADAR sensor with a rectangular beamformer) may be located on the same PCB 514 or may be located remotely and connected to the PCB 514 via an electronic conductor or a wireless transceiver.

[0098] The interface connector 510 may provide power and universal asynchronous receiver / transmitter (UART) functionality between the RFCA module 500 and the aircraft 200. The RFCA module 500 is preferably compact and partially vehicle and sensor agnostic, and is configured to act as a retrofit to an existing flight control system 306 and / or flight control mechanism(s) 304 of the aircraft 200 via the interface connector 510. In operation, the sensors 210 provide input measurements to the microcontroller 508. The microcontroller 508 is configured to execute one or more software programs (e.g., navigation, altitude hold, landing assistance, and crash protection software / functions). The microcontroller 508 outputs commands to the flight control system 306 of the aircraft 200. The RFCA module 500 is advantageous in that it provides a reliable and flexible architecture that can operate as a single module for capturing raw data (e.g., from one or more sensors 210), signal processing, and detection algorithms. Data communications between the RFCA module 500 and the aircraft 200 may be facilitated via a single bidirectional communication channel that may be configurable through, for example, a MAV Link application program interface (API).

[0099] The RFCA module 500 may be installed on the aircraft 200 in a downward-facing configuration, a forward-facing configuration, or a combination thereof. In the downward-facing configuration, the RFCA module 500 may facilitate altitude keeping and landing assistance functions (e.g., via a landing assistance module), as shown in FIG. 2c. For example, the RFCA module 500 may command the aircraft 200 to maintain a predetermined altitude (above the ground) relative to a detected obstruction below, via the altitude keeping function, or command the aircraft 200 to perform a landing maneuver to avoid a detected obstruction below, in the landing assistance function. In the forward-facing configuration, the RFCA module 500 provides collision protection by commanding a forward flight maneuver (e.g., acting as a brake) to avoid a collision with a detected obstruction, as shown in FIG. 2d. Preliminary testing indicates that the RFCA module 500 provides a range of 1 to 10 meters, speeds up to 5 m / s, an update rate of 40 Hz, a field of view (FOV) of 70 degrees (in forward-facing configuration) and a field of view of 60 degrees (in downward-facing configuration), and a resolution of 8 cm for the nearest object detected, whether stationary or moving.

[0100] The RFCA module 500 is preferably compact and lightweight, thereby minimizing loads and burdens on the aircraft 200. For example, the RFCA module 500 may have a length (L) between 10 mm and 50 mm, more preferably between 20 mm and 40 mm, and most preferably about 30 mm, while the width (W) may be between 10 mm and 40 mm, more preferably between 15 mm and 30 mm, and most preferably about 20 mm, and the height (H) may be between 1 mm and 10 mm, more preferably between 3 mm and 7 mm, and most preferably about 5 mm. For example, an RFCA module 500 having dimensions of 20×30×5 mm has been shown to have a weight of less than 25 grams and a power consumption of less than 1 watt (W).

[0101] Three-area collision protection In certain aspects, the processor 340 of the dynamic collision avoidance system 302 may perform the three-domain collision protection function with pilot priority features. However, in certain aspects, a separate module may be provided that includes circuitry to facilitate the three-domain collision protection function. The three-domain collision protection function provides sensor and aircraft-agnostic techniques to prevent inadvertent collision of the aircraft 200 with objects within its environment / range. In other words, the three-domain collision protection function provides the aircraft 200 with “detect and avoid” and / or “sense and avoid” capabilities that are essential for the navigation of the aircraft, whether through a pilot or flown autonomously.

[0102] An advantage of the aircraft 200 employing a three-zone collision protection function (e.g., via the three-zone collision protection algorithm 600) is that it can utilize low-cost and low fidelity distance measurements to provide an effective collision protection system for actively piloted aircraft 100. The three-zone collision protection function can also scale sensor inputs from one or more sensors 210 while maintaining a sensor-agnostic and aircraft-agnostic character in its design, thereby allowing the three-zone collision protection function to be employed on virtually any aircraft.

[0103] An exemplary three zone collision protection function may be facilitated by using a three zone collision protection algorithm 600 shown in FIG. 6a. As shown, inputs to the three zone collision protection algorithm 600 from one or more sensors 210, a human pilot, or an autopilot are received at step 602. The inputs may include a range rate prediction, a range prediction, and an input pilot command stream (i.e., pilot commands). The input pilot command stream may be a command stream from a human operator or an autopilot. Using these three inputs, the three zone collision protection algorithm 600 determines at step 604 a zone in which the target is currently located from a plurality of zones in the physical space. As will be described, the zone is an area or portion of the space typically defined by the field of view of the sensor 210 between the vehicle 200 and the operating range (e.g., maximum operating distance) of the sensor 210. In one embodiment, the three zones are identified as an incoming zone 622, a critical zone 620, or a panic zone 618. The three regions may be identified, for example, using the formula provided in step 606 .

[0104] In another embodiment, as shown in FIG. 6b, three regions may be identified based on operator-determined sensor distance thresholds (e.g., first and second distance thresholds 624, 626). The thresholds define the distance boundaries. As shown, the first region (incoming region 622) spans the area between the sensor maximum distance and the incoming threshold 626 (e.g., first threshold / distance), the second region (critical region) spans the area between the incoming threshold 626 and the panic threshold 624, and the third region (panic region 618) spans the area between the panic threshold 624 (e.g., second threshold / distance) and the vehicle itself. As can be understood, the sensor maximum distance may refer to the maximum distance at which a given sensor at that point can detect an object / target with adequate reliability. As shown, the shape of each region may be determined depending on the shape of the field of view (FOV). The sensor distance thresholds 624, 626 may be received by the aircraft 200 from the operator. For example, if the target is located in either the incoming or critical region, then a rate limit curve is calculated in the range, range and rate domains using the distance threshold and range rate limit of the region set by the operator. Based on these determinations, in step 606, the three-region collision protection algorithm 600 adjusts the control inputs (e.g., Rate set and / or K p ) from step 606. The three-domain collision protection algorithm 600 then sets the Rate set and / or K pThe control inputs are input to a Proportional-Derivative (PD) controller at step 608, which outputs control data. At step 610, the three domain collision protection algorithm 600 converts the control data from step 608 (i.e., dimensionless, ranging from -1 to 1) into a control command stream (i.e., ranging from 1000 to 2000 in units of the pilot commands input from step 602, such as PWM pulse lengths). At step 612, the three domain collision protection algorithm 600 compares the control command stream from step 610 to the pilot command stream (i.e., pilot commands, i.e., the input pilot command stream of step 602) to determine whether the pilot commands are safe. If the commands reduce the vehicle-to-target distance or increase the vehicle rate above the rate limit set at 606 (e.g., Rate set ), the pilot command is presumed to be unsafe. If the pilot command is determined to be unsafe (i.e., unsafe), then in step 614 the three zone collision protection algorithm 600 outputs the input pilot command stream from step 602. If the pilot command is determined to be safe, then in step 616 the three zone collision protection algorithm 600 outputs the input pilot command stream from step 602.

[0105] FIG. 6b illustrates an exemplary embodiment of three regions: panic region 618, danger region 620, and incoming region 622. Aircraft 200 may have the field of view shown in FIG. 6b. In this embodiment, the field of view represents the field of view of the sensors up to the sensor maximum distance. The field of view may be divided into the three regions based on operator threshold inputs, such as panic threshold 624 and incoming threshold 626. As such, control of aircraft 200 may be uniquely limited as aircraft 200 enters each of the three regions. By way of example, maximum speed, acceleration, and / or rate limits of vehicle 200 may be differently limited as aircraft 200 enters each of the three regions 618, 620, 622. Once aircraft 200 is located in panic region 618, the system may control aircraft 200 to slow down. As aircraft 200 enters danger region 620, aircraft 200 may slow down or brake further. Finally, when aircraft 200 is positioned within incoming region 622 (e.g., aircraft 200 is in the closest region toward a visible obstacle), a command to apply brakes may be issued and / or a forward command input by the operator may be ignored to override any control commands that steer aircraft 200 forward. These and various other examples of limiting commands / control will be apparent to those of skill in the art in light of this disclosure.

[0106] Landing Assistance Module The landing assistance module includes circuitry for autonomously landing the aircraft 200 with nominal input from the operator by autonomously controlling the flight control mechanisms 304 and / or the throttle during the landing operation. However, in certain aspects, the processor 340 of the dynamic collision avoidance system 302 may perform the landing assistance functions. The landing assistance module benefits from sensor feedback to inhibit the aircraft 200 from landing on an obstruction and / or landing in a position that may result in a collision or other hazard to the aircraft 200 (e.g., due to an obstacle). The landing assistance module uses the sensor input during the landing maneuver to close the loop and ensure a smooth and safe landing. More specifically, the landing assistance module employs input from one or more sensors 210 that can measure distances to multiple targets below the aircraft 200. For example, the one or more sensors 210 can be RADAR, LiDAR, stereo vision (via two or more cameras), etc.

[0107] The landing assistance system provides closed-loop control and safe landing confirmation to the flight control system 306 of the aircraft 200 without the need for beacons or fiducial overseers. By surveying landing locations (e.g., at the target 102) and calculating a confidence value based on a target filter (described below), the aircraft 200 can determine whether an area below the aircraft 200 (e.g., via the dynamic collision avoidance system 302 / RFCA module 500) can be used for closed-loop autonomous landing. In addition to identifying and / or confirming safe landing areas, the landing assistance module can also generate throttle and / or control commands for the flight control mechanisms 304 to smoothly land the aircraft 200.

[0108] The landing assistance module maintains safe operation of the aircraft 200 through altitude clearance and closed-loop throttle commands. The landing assistance module measures the altitude of the aircraft 200 through one or more sensors 210 and performs closed-loop throttle control of the aircraft 200 to provide landing assistance. In the event that an obstacle is detected (e.g., by the RFCA module 500 in a downward-facing configuration or another sensor 210), a postponement of the landing is performed, whereby the aircraft 200 can abort its landing and / or identify a new landing area. Furthermore, distance measurements are filtered to first determine a landing target, then generate closed-loop throttle control / control commands, and finally "postpone" (e.g., abort) an autonomous landing that does not meet landing parameters (e.g., confidence, maximum distance, and descent rate). The RFCA module 500 may subsequently identify a new landing area based on the sensor measurements, or perform maneuvers other than hovering when the landing area is estimated to be unsafe.

[0109] 7 illustrates an example landing assistance flow chart 700 for the landing assistance function of the landing assistance module. As shown, inputs to the landing assistance module are received at step 702. The inputs may include a range array and landing parameters, which may be generated by either a human operator or an autopilot.

[0110] In step 704, the landing target filter determines the number of targets from the range array. In step 706, the landing assistance module determines whether the number of targets is greater than one. If the number of targets is greater than one, the landing assistance module proceeds to step 708, otherwise, the landing assistance module proceeds to step 710. In step 708, the landing assistance module sets the target equal to min(distance_array) and proceeds to step 710. In step 710, the landing assistance module determines whether the target meets the landing parameters. If the target meets the landing parameters, the landing assistance module proceeds to step 712, otherwise, the landing assistance module proceeds to step 716. In step 712, the landing assistance module sets LANDING_STATUS to FALSE and proceeds to step 714. In step 714, the landing assistance module outputs LANDING_STATUS to, for example, the flight control system 306. In step 716, the landing assistance module sets LANDING_STATUS to TRUE and proceeds to step 718. In step 718, the landing assist module employs a proportional-integral-derivative (PID) controller to generate a control signal and proceeds to step 720. In step 720, the landing assist module converts the control signal from step 718 into units of Throttle_Cmd. In step 722, the landing assist module outputs Landing_Status and Throttle_Cmd from step 720, for example, to the flight control system 306. The vehicle's flight control system can then query the Landing_Status and take appropriate action to apply Throttle_Cmd to the appropriate control loops.

[0111] Default outputs from the landing assist module include, for example: altitude_clearance, pilot_thrust_command, and landing_status. However, the outputs can be customized to meet specific needs. Custom outputs can include, for example: altitude_relative, target_VZ (velocity in Z direction). Landing assist module parameters can be configured through the API.

[0112] Target Filtering Module A problem with RADAR sensors is that due to their ability to detect objects over large areas, the output is usually scattered and chaotic data. As a result, it can be difficult to analyze the data returned from a RADAR sensor, especially for a navigation decision process. In operation, a RADAR sensor may measure the five most prominent objects within its line of sight and output both the relative distance of the objects and the size that the RADAR is detecting. These values ​​are usually output in descending order of magnitude, but this method of feedback is inappropriate for collision avoidance and autonomous flight because in a dynamic environment, obstacles (as seen by the RADAR) constantly surpass each other in magnitude and the reported distances are interchanged, thus resulting in large jumps in feedback that severely impact the closed-loop controller.

[0113] The solution is to employ a target filtering module that includes circuitry implementing a target filtering algorithm to detect real objects versus noise and track the objects such that a clean feedback signal is returned to the aircraft 200. The target filtering module implements a target filtering algorithm to prevent jumps, smooth the output, and report a confidence value. The confidence value may be used to prevent false positives in a collision avoidance system. Target filtering and optimization of the five range, five magnitude RADAR device thus allows for identifying and tracking obstructions using data obtained from the RADAR sensor. The target filtering module may thus address this problem by providing filtering and optimization of the five range, five magnitude RADAR sensor payload for the aircraft 200, allowing for five objects to be tracked simultaneously.

[0114] The ability to perform object tracking enhances both autonomous navigation and collision avoidance. In certain aspects, the processor 340 of the dynamic collision avoidance system 302 may perform the disclosed target filtering function. Thus, the target filtering module converts unstable and noisy RADAR measurements into clean signals that can be utilized for collision avoidance as well as autonomous flight feedback. The target filtering module emphasizes the advantages of the RADAR module while attenuating its shortcomings to optimize the performance of the aircraft 200. Similar solutions can be derived by minor modifications of the current algorithms. There is some flexibility in the selection of sub-functions (i.e., type of digital filter, linear vs. polynomial curve fitting, Gaussian vs. binomial predictive distribution, etc.).

[0115] 8 illustrates an input and output diagram 800 of an exemplary target filter module. The exemplary target filter module includes circuitry for combining sensor data from one or more sensors 210 using a target filter algorithm 812 to generate a filtered range with a corresponding confidence value. The one or more sensors 210 may include a RADAR and an altimeter for collecting altitude measurements 802, RADAR magnitude 804, and RADAR distance 806. Using signal property extraction techniques 810, the target filter algorithm 812 can detect anomalous measurements from the RADAR and eliminate erroneous data points. Range assignment and confidence values ​​are calculated using a weighted average of the signal properties, RADAR magnitude, and predicted values ​​of ground noise.

[0116] FIG. 9 shows an example flow chart 900 for providing a target filtering function using the target filtering module. Here, Std_Dev is the standard deviation of the latest 20 points of the trace through a linear regression of these 20 points. Min_Diff is the minimum difference between the latest distance of the trace and the assigned distance from the incoming data. Iteration_Counter (iteration count or counter) is the number of consecutive extrapolations. Finally, the confidence value is the confidence of the distance being the actual target. The confidence is calculated using a weighted sum of the magnitude and Std_Dev.

[0117] In step 902, the target filtering module receives the first of five new distances and magnitudes from the RADAR for the five most prominent objects within line of sight and proceeds to step 904. In step 904, the target filtering module determines whether the magnitudes have been met. If the magnitudes have been met in step 904, the target filtering module sets the distances to a previously known good value in step 906 and proceeds to step 908. If the magnitudes have not been met in step 904, the target filtering module continues to step 908. In step 908, the target filtering module calculates Std_Dev and proceeds to step 910. In step 910, the target filtering module calculates and / or identifies the new distance points that best fit the current trace and proceeds to step 912. In step 912, the target filtering module calculates Min_Diff and proceeds to step 914. In step 914, the target filtering module determines whether four conditions have been met. The four conditions are: (1) whether Min_Diff is greater than 3.5 times Std_Dev, (2) whether Min_Diff is greater than 0.4, (3) whether Std_Dev is less than 0.2, and (4) whether iteration_counter is less than 15. In step 914, if each of the four conditions is met, the target filtering module proceeds to step 916;

[0118] In step 916, the target filtering module calculates new filtered distance points using linear regression and proceeds to step 918. In step 918, the target filtering module increments the iteration_counter and proceeds to step 924.

[0119] In step 920, the target filtering module assigns the best data point from the RADAR to filter the output distances and proceeds to step 922. In step 922, the target filtering module removes the assigned distances from distances potentially used in the trace and proceeds to step 924.

[0120] In step 924, the target filtering module calculates the confidence and low pass using a critically damped low pass filter (LPF) and proceeds to step 926. The confidence and low pass values ​​may be calculated using a weighted average of statistical terms derived from the input distance signal (e.g., signal mean, standard deviation, and magnitude). The weighting of the terms in the confidence calculation may be operator determined (i.e., operator specified) and adjusted (e.g., by the operator) for the desired filter performance. In step 926, the target filtering module repeats the above (starting at step 902) for each of distances 2-5, and proceeds to step 928 when all distances 1-5 are completed. In step 928, the target filtering module returns the filtered distance and confidence values.

[0121] In some embodiments, the confidence value and low pass value may be determined based on the environment in which the aircraft 200 typically operates. For example, in an operating environment in which snow is common, snow reflectivity may be considered a common cause of detected levels of noise. As such, the system may set the confidence value and / or low pass value accordingly to adjust the acceptable noise level detected by the RADAR. Similarly, machine learning algorithms may be applied using the target filtering capabilities of the system just described.

[0122] The systems described herein may also include client devices. The client device may include any device operated by an operator to initialize, manage, monitor, control, or otherwise interact with the navigation system or the autonomous vehicle. This may include a desktop computer, a laptop computer, a network computer, a tablet, or any other computing device that may be included in the systems contemplated herein. The client device may include an operator interface. The operator interface may include a graphical user interface (GUI), a text or command line interface, a voice control interface, and / or a gesture-based interface for controlling the operation of the navigation system or the autonomous vehicle. The operator interface may be maintained, for example, by an application running locally on one of the client devices that receives data and status information from the navigation system or the autonomous vehicle. The operator interface may generate an appropriate display on the client device for operator interaction. For example, the operator interface may include a display that displays a view from a camera in the optical system in real time or other data from other sensors in the navigation system. In other embodiments, the operator interface may be serviced and presented remotely on one of the client devices. For example, the navigation system or autonomous vehicle may include a web server that provides information, such as through one or more web pages that may be displayed within a web browser or similar client executing on one of the client devices. In one aspect, the operator interface may include a voice-controlled interface that receives spoken commands from the operator and / or provides spoken feedback to the operator.

[0123] Although the above-described systems have been described primarily as applied to air vehicles, those skilled in the art will appreciate that such systems, methods, and techniques may be employed with other technologies, such as automobiles, warehouse equipment, construction equipment, cranes, power wheelchairs, airport equipment, etc.

[0124] The above-mentioned systems, devices, methods, processes, etc. may be implemented in hardware, software, or any combination thereof suitable for the control, data acquisition, and data processing described herein. This includes implementation within one or more microprocessors, microcontrollers, embedded microcontrollers, programmable digital signal processors, or other programmable devices or processing circuits, along with internal and / or external memory. This may additionally or alternatively include one or more application specific integrated circuits, programmable gate arrays, programmable array logic components, or any other device or devices that may be configured to process electronic signals. It will be further understood that implementations of the above-mentioned processes or devices may include computer executable code generated using a structured programming language such as C (including assembly language, hardware description languages, and database programming languages ​​and techniques), an object-oriented programming language such as C++, or any other high-level or low-level programming language, that may be stored, compiled, or translated to be executed in one of the above-mentioned devices, as well as heterogeneous combinations of processors, processor architectures, or combinations of different hardware and software. At the same time, processing may be distributed across devices such as the various systems described above, or all of its functionality may be integrated into a dedicated stand-alone device, and all such variations and combinations are intended to fall within the scope of this disclosure.

[0125] The embodiments disclosed herein may include a computer program product comprising computer executable or computer usable code that, when executed on one or more computing devices, performs any and / or all of the steps of the control system described above. The code may be stored in a non-transitory manner in a computer memory. The computer memory may be a memory (such as a random access memory coupled to a processor) from which the program executes, or a storage device such as a disk drive, flash memory, or any other optical, electromagnetic, magnetic, infrared, or other device or combination of devices. In another aspect, any of the above-mentioned systems and methods may be embodied in any suitable transmission or propagation medium that carries the computer executable code and / or any input or output therefrom.

[0126] The method steps of the embodiments described herein are intended to include any suitable manner of causing one or more other parties or entities to perform the steps consistent with the patentability of the claims below, unless a different meaning is clearly stated or is otherwise apparent from the context. Such parties or entities need not be under the direction or control of any other party or entity, nor need they be located within any particular jurisdiction.

[0127] Any patents, patent publications, or articles cited herein are incorporated herein by reference in their entirety. It will be understood that the above-mentioned methods and systems are described in a non-limiting manner by way of example. Many variations, additions, omissions, and other modifications will be apparent to those skilled in the art. Furthermore, the order or representation of the method steps described above and in the accompanying drawings is not intended to require performing the recited steps in this order unless a particular order is expressly stated or is otherwise clear from the context. Thus, while specific embodiments have been illustrated and described, it will be apparent to those skilled in the art that various changes and modifications in form and detail may be made without departing from the spirit and scope of this disclosure, and that such changes and modifications are intended to form part of the present invention as defined by the following claims, which claims are to be interpreted in the broadest sense permitted by law. The present invention also includes the following aspects. (Aspect 1) 1. A method for providing crash protection in an aircraft (100, 200), comprising: receiving a sensor input from a sensor (210, 323) positioned on the aircraft (100, 200) and operatively connected to a processor (314, 340), the sensor (210, 323) configured to identify an obstacle (122) within a field of view; receiving a pilot command stream from a pilot; identifying an obstacle (122) within said field of view based at least in part on said sensor input; determining a region (618, 620, 622) in which the obstacle (122) is located from a plurality of regions (618, 620, 622) within the field of view, the region (618, 620, 622) being determined at least in part based on the sensor input; setting a control input in response to the area (618, 620, 622) determined in the determining step; inputting the control input into a proportional-derivative (PD) controller to generate control data; generating, via said processor (314, 340), a control command stream in response to said control data; and and comparing, via the processor (314, 340), the control command stream to the pilot command stream to determine whether the pilot command stream from the pilot is safe. (Aspect 2) 2. The method of claim 1, wherein when the pilot command stream from the pilot is determined to be unsafe, the control command stream is communicated to a flight controller (206, 306) of the aircraft (100, 200) in place of the pilot command stream. (Aspect 3) 3. The method of claim 1 or 2, wherein the pilot command stream is communicated to a flight controller (206, 306) of the aircraft (100, 200) when the pilot command stream from the pilot is determined to be safe. (Aspect 4) 4. The method of any of aspects 1-3, wherein the sensor input comprises a range rate prediction or a range prediction. (Aspect 5) 5. The method of any one of aspects 1 to 4, wherein the pilot command stream is from a human pilot. (Aspect 6) 6. The method of any one of aspects 1 to 5, wherein the pilot command stream is from an autopilot. (Aspect 7) 7. The method of any of aspects 1 to 6, wherein the plurality of regions (618, 620, 622) comprises a first region (618, 620, 622), a second region (618, 620, 622), and a third region (618, 620, 622). (Aspect 8) 8. The method of embodiment 7, wherein the first region (618, 620, 622) is defined as an area between a maximum sensor (210, 323) distance and a first threshold. (Aspect 9) 9. The method of any one of aspects 7 to 8, wherein the second region (618, 620, 622) is defined as an area between a first threshold and a second threshold. (Aspect 10) 10. The method of any of aspects 7 to 9, wherein the third region (618, 620, 622) is defined as an area between a second threshold and the aircraft (100, 200). (Aspect 11) 11. The method of any of aspects 1 to 10, wherein the pilot command stream is determined to be unsafe if the pilot command stream can be interpreted by the processor (314, 340) as attempting to (1) reduce the distance between the aircraft (100, 200) and the obstacle (122), or (2) increase the rate of the aircraft (100, 200) above a rate limit set by the control data. (Aspect 12) 12. The method of any of aspects 1-11, further comprising receiving a pilot priority command from the pilot, the pilot priority command taking priority over the control command stream. (Aspect 13) 1. A navigation system for providing collision protection in an aircraft (100, 200), comprising: a sensor (210, 323) coupled to the aircraft (100, 200) and configured to identify an obstacle (122) within visual range; a processor (314, 340) operatively connected to the sensor (210, 323) and the memory device; the sensor (210, 323) is configured to receive a pilot command stream from a pilot; and configured to identify obstacles (122) within said field of view based at least in part on sensor input from said sensors (210, 323); determining a region (618, 620, 622) in which the obstacle (122) is located from a plurality of regions (618, 620, 622) within the field of view, the region (618, 620, 622) being determined based at least in part on the sensor input; configured to set a control input in response to the area (618, 620, 622) determined in the determining step; configured to input the control input to a proportional-derivative (PD) controller to generate control data; configured to generate, via said processor (314, 340), a control command stream in response to said control data; A navigation system configured to, via the processor (314, 340), compare the control command stream with the pilot command stream to determine whether the pilot command stream from the pilot is safe. (Aspect 14) The navigation system of aspect 13, wherein when the pilot command stream from the pilot is determined to be unsafe, the control command stream is communicated to a flight controller (206, 306) of the aircraft (100, 200) in place of the pilot command stream. (Aspect 15) The navigation system of any one of claims 13 to 14, wherein the pilot command stream is communicated to a flight controller (206, 306) of the aircraft (100, 200) when the pilot command stream from the pilot is determined to be safe. (Aspect 16) 16. The navigation system of any of aspects 13 to 15, wherein the sensor input includes a range rate prediction or a range prediction. (Aspect 17) 17. The navigation system of any of aspects 13 to 16, wherein the pilot command stream is from a human pilot. (Aspect 18) 18. A navigation system according to any one of aspects 13 to 17, wherein the pilot command stream is from an autopilot. (Aspect 19) 19. The navigation system of any of aspects 13 to 18, wherein the plurality of regions (618, 620, 622) includes a first region (618, 620, 622), a second region (618, 620, 622), and a third region (618, 620, 622). (Aspect 20) 20. The navigation system of embodiment 19, wherein the first region (618, 620, 622) is defined as an area between a sensor maximum distance and a first threshold. (Aspect 21) The navigation system of aspect 19 or 20, wherein the third region (618, 620, 622) is defined as an area between a second threshold and the aircraft (100, 200). (Aspect 22) 22. The navigation system of embodiment 21, wherein the second region (618, 620, 622) is defined as an area between a first threshold and the second threshold. (Aspect 23) The navigation system of aspect 13, wherein the pilot command stream is determined to be unsafe if the pilot command stream can be interpreted by the processor (314, 340) as attempting to (1) reduce the distance between the aircraft (100, 200) and the obstacle (122), or (2) increase the rate of the aircraft (100, 200) above a rate limit set by the control data. (Aspect 24) 24. The navigation system of any of aspects 13 to 23, wherein the processor (314, 340) is configured to receive a pilot priority command from the pilot that takes priority over the control command stream. (Aspect 25) 25. The navigation system of any of aspects 13 to 24, wherein the aircraft (100, 200) is a vertical take-off and landing (VTOL) aircraft (100, 200). (Aspect 26) 26. The navigation system of any of aspects 13 to 25, further comprising a landing assistance module that instructs the aircraft (100, 200) to perform landing maneuvers to avoid obstacles detected below the aircraft (100, 200). (Aspect 27) 27. The navigation system of any of aspects 13 to 26, wherein the processor (314, 340) is configured to perform a target filtering operation. (Aspect 28) The target filtering operation comprises: receiving range and size data for an obstacle (122) within line of sight of the aircraft (100, 200) from a RADAR system; determining, via said processor (314, 340), whether said magnitude has been met based at least in part on said distance and magnitude data; calculating, via said processor (314, 340), a standard deviation of at least a portion of a trace reflecting said distance and magnitude data over time; determining, via said processor (314, 340), new distance points for said trace; calculating, via said processor (314, 340), a minimum difference between said new distance points for said trace and an assigned distance from the incoming data; and 28. The navigation system of aspect 27, comprising calculating, via the processor (314, 340), a confidence and low pass value by a critically damped low pass filter (LPF). (Aspect 29) 30. The navigation system of embodiment 28, wherein the confidence and low pass values ​​are calculated using a weighted average of statistical terms derived from the signal mean, standard deviation, and magnitude. (Aspect 30) 30. The navigation system of embodiment 29, wherein the weighted average is defined by an operator for a desired filter performance. (Aspect 31) 1. A method for providing precision target filtering in an aircraft (100, 200), comprising: receiving range and size data for an obstacle (122) within line of sight of said aircraft (100, 200) from a RADAR system; determining whether the magnitude has been met based at least in part on the distance and magnitude data via a processor (314, 340), wherein if the magnitude has been met, the processor is configured to set the distance to a known appropriate value; calculating, via said processor (314, 340), a standard deviation of at least a portion of a trace reflecting said distance and magnitude data over time; determining, via said processor (314, 340), new distance points for said trace; calculating, via said processor (314, 340), a minimum difference between said new distance points for said trace and an assigned distance from the incoming data; determining via said processor (314, 340) whether each of a plurality of conditions is satisfied, wherein if one or more of said plurality of conditions is not satisfied, said processor (314, 340) is configured to calculate new filtered distance points using linear regression; incrementing the iteration counter, and via the processor (314, 340), calculating a confidence and a low pass value with a critically damped low pass filter (LPF). (Aspect 32) 32. The method of claim 31, wherein each of the steps recited in aspect 31 is performed for each of a predetermined number of obstacles (122) within line of sight of the aircraft (100, 200). (Aspect 33) 33. The method of claim 32, wherein the predetermined number of obstacles includes five most prominent obstacles (122) within line of sight of the aircraft (100, 200). (Aspect 34) 33. The method of any one of embodiments 31 to 32, wherein the standard deviation is the standard deviation of the most recent 20 points of the trace through a 20-point linear regression. (Aspect 35) 35. The method of embodiment 34, wherein the minimum difference is the difference between a latest distance of the trace and an assigned distance from incoming data. (Aspect 36) 36. The method of claim 35, wherein the plurality of conditions include (1) whether the smallest difference is greater than 3.5 times the standard deviation, and (2) whether the smallest difference is greater than 0.4. (Aspect 37) 37. The method of embodiment 36, wherein the plurality of conditions further includes: (1) whether the standard deviation is less than 0.2; and (2) whether the iteration counter is less than 15. (Aspect 38) 38. The method of any of embodiments 31 to 37, wherein the confidence and low pass values ​​are calculated using a weighted average of statistical terms derived from signal mean, standard deviation, and magnitude. (Aspect 39) 39. The method of any of embodiments 31-38, wherein the weighted average is defined by an operator with respect to a desired filter performance.

Claims

1. A method for providing collision avoidance in an aircraft (200), comprising: receiving a sensor input from a sensor (210) positioned on the aircraft (200) and operatively connected to a processor (314, 340), the sensor (210) configured to identify an obstacle (122) within a field of view; receiving a pilot command stream from a pilot; identifying an obstacle (122) within said field of view based at least in part on said sensor input; determining a region (618, 620, 622) in which the obstacle (122) is located from a plurality of regions (618, 620, 622) within the field of view, the region (618, 620, 622) being determined at least in part based on the sensor input; setting a control input in response to the area (618, 620, 622) determined in said determining; inputting the control input into a proportional-derivative (PD) controller to generate control data; generating, via said processor (314, 340), a control command stream in response to said control data; comparing, via the processor (314, 340), the control command stream with the pilot command stream to determine whether the pilot command stream from the pilot is safe; and communicating the control command stream to a flight controller (206, 306) of the aircraft (200) in place of the pilot command stream when the pilot command stream from the pilot is determined to be unsafe.

2. 2. The method of claim 1, wherein the pilot command stream is communicated to the flight controller (206, 306) of the aircraft (200) when the pilot command stream from the pilot is determined to be safe.

3. The method of claim 1 or 2, wherein the sensor input comprises a range rate prediction or a range prediction.

4. The method of claim 1 , wherein the pilot command stream is from a human pilot.

5. 4. The method of claim 1, wherein the pilot command stream is from an autopilot.

6. 6. The method of claim 1, wherein the plurality of regions includes a first region, a second region, and a third region, the first region being defined as a region between a sensor maximum distance and a first threshold, the second region being defined as a region between the first threshold and a second threshold, and the third region being defined as a region between the second threshold and the aircraft.

7. 7. The method of claim 1, wherein the pilot command stream is determined to be unsafe if the pilot command stream can be interpreted by the processor as attempting to (1) reduce a distance between the aircraft (200) and the obstacle (122), or (2) increase the rate of the aircraft (200) above a rate limit set by the control data.

8. 1. A navigation system for providing collision avoidance in an aircraft (200), comprising: a sensor (210) coupled to the aircraft (200) and configured to identify obstacles (122) within visual range; a processor (314, 340) operatively connected to the sensor (210) and a memory device; the processor (314, 340) configured to receive a pilot command stream from a pilot; and configured to identify an obstacle (122) within the field of view based at least in part on sensor input from the sensor (210); determining a region (618, 620, 622) in which the obstacle (122) is located from a plurality of regions (618, 620, 622) within the field of view, the region (618, 620, 622) being determined based at least in part on the sensor input; configured to set a control input in response to the determined area (618, 620, 622); configured to input the control input to a proportional-derivative (PD) controller to generate control data; configured to generate, via said processor (314, 340), a control command stream in response to said control data; configured, via the processor (314, 340), to compare the control command stream with the pilot command stream to determine whether the pilot command stream from the pilot is safe; When the pilot command stream from the pilot is determined to be unsafe, the control command stream is communicated to a flight controller (206, 306) of the aircraft (200) in place of the pilot command stream.

9. 9. The navigation system of claim 8, wherein the pilot command stream is communicated to the flight controller (206, 306) of the aircraft (200) when the pilot command stream from the pilot is determined to be safe.

10. 10. A navigation system according to claim 8 or 9, wherein the sensor input comprises a range rate prediction or a range prediction.

11. A navigation system according to any one of claims 8 to 10, wherein the pilot command stream is from a human pilot.

12. A navigation system according to any one of claims 8 to 10, wherein the pilot command stream is from an autopilot.

Citation Information

Patent Citations

  • JP1973040240A

  • Obstacle avoiding system for flying object and manned flying object

    JP2003127997A

  • Obstacle avoidance system for stabilized aerial vehicle and method of controlling same

    WO2016008026A1