CN108681447A - A kind of Automatic parameter selection method of robot obstacle-avoiding - Google Patents
A kind of Automatic parameter selection method of robot obstacle-avoiding Download PDFInfo
- Publication number
- CN108681447A CN108681447A CN201810468201.2A CN201810468201A CN108681447A CN 108681447 A CN108681447 A CN 108681447A CN 201810468201 A CN201810468201 A CN 201810468201A CN 108681447 A CN108681447 A CN 108681447A
- Authority
- CN
- China
- Prior art keywords
- event
- robot
- monitor
- parameter
- parameter selection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/20—Software design
- G06F8/24—Object-oriented
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
Abstract
The present invention relates to a kind of Automatic parameter selection methods of robot obstacle-avoiding：Step 1: the foundation and execution of run time verification model：Run time verification model, including variable declarations, definition event, description attribute and the several parts of self-defined user program are established according to the grammer of JavaMOP；Monitor real time monitoring is ultimately produced, realizes on line real-time monitoring；Step 2: Automatic parameter selection.(1) definition of the event recorder of monitor identification events from specification, and in logging program control flow all events possibility entry；(2) event that the filter matching of monitor defines, but prevent every other event；(3) parameter of capturing events is modified, and new parameter may return to the source finally continued.The present invention can be offline by Automatic parameter selection method, real-time undated parameter, play an important roll for behavior restraint of mobile robot in the case of environment is uncertain.
Description
Technical field
The present invention relates to a kind of Automatic parameter selection methods of robot obstacle-avoiding, belong to robot security's technical field.
Background technology
Robot completes heavy and dangerous work in production and life instead of the mankind, effectively reduces the labor of the mankind
Fatigue resistance.With industrial robot, fire-fighting robot, tour guide, service robot, the application of the mobile robots such as amusement robot
More and more extensive, people are more and more to the concern of its safety.Avoid collision be most important one a bit.
Many methods have had been proposed about avoidance problem.When barrier is static or robot movement
Path can be known in advance or assess in dynamic environment to design barrier-avoiding method.But in fact, what robot worked
Environment is uncertain, and in other words, the quantity of dynamic barrier, position and motion path are all unknown.In such case
Under, also occur various intelligent barrier avoiding algorithms both at home and abroad.Mobile robot can control robot using fuzzy logic method
Movement in a stationary obstruction region keeps the desired distance between robot and barrier using virtual ARTIFICIAL FORCE field,
And the uncertainty for the task of avoiding collision also is compensated using neural network.In addition, it is also proposed that many sensor-based
The method that Real-time Decision can be made.A kind of new Real-time Obstacle Avoidance Method VFH is to use planar perspective flute card Gifid as ring
Border model builds obstacle density in sector, then moves machine by ultrasonic sensor sample range data real-time update
People can break the barriers from the low nonpolar sector selection suitable path of obstacle density.It is identified using sensor and avoids hindering
It is also a kind of method being often used to hinder object.Ultrasonic wave and infrared sensor may be used for detection and barrier and its position away from
From, it can be by surrounding center rotation searching target direction, then straight trip reaches target.It environment can also realize independently indoors
The air navigation aid of the view-based access control model of mobile robot.
Above-mentioned algorithm is intended to avoid the collision in design, but can also have many problems in force.For example, sensor
The transmission time of data cannot be ignored.If the order of control system can not be executed timely, sensing data can
It can not can correctly reflect the virtual condition of robot at that time.The appearance of these situations will bring many problems.Robot
False judgment may bring a series of behavior of danger, and can be had an impact to operator and robot itself.In addition, should
Some special parameters of program may can not be fixed in the development phase, need to change with the variation of environment in true environment
Become, or real-time update is carried out according to the experience of engineer.
In order to capture and avoid above-mentioned dangerous problem, it is proposed that robot run time verification (RV) method is referred to
In robot system, to monitor behavior of the robot in uncertain environment.In addition, we also proposed a kind of dynamic update ginseng
The program of robot itself is familiar with without operating personnel and is changed to several methods.
Run time verification is a kind of lightweight authentication technology, user-defined dynamic when an attribute is breached or verifies
Work will be triggered.It has been applied to every field, such as wireless sensor application, vehicle bus system, and weblication,
Java programs, c program and medical physics system.We have selected a run time verification tool JavaMOP in the present invention, it
It is the example of a MOP (being programmed towards monitoring), the behavior of mobile robot is monitored for generating a monitor.MOP is
One general monitoring frame, it integrates specification and realization by checking the former with the latter at runtime.It is to insert
The form of part supports many different logics, and carrys out quick access monitor using identical Index Algorithm.
Invention content
It is in verification at runtime the purpose of the present invention is to provide a kind of Automatic parameter selection method of robot obstacle-avoiding
The method for carrying out dynamic parameter selection can apply the dynamic real-time update of the parameter in uncertain scene, be not necessarily to operating personnel
It is familiar with and changes the program of robot itself.We pass through the method monitored when operation and monitor in real time, agility reaction, and by certainly
Dynamic adjusting parameter carrys out the danger zone data of real-time update robot when driving, to achieve the purpose that accurate avoidance.
A kind of Automatic parameter selection method of robot obstacle-avoiding of the present invention, is as follows：
Step 1: the foundation and execution of run time verification model
Run time verification model, including variable declarations, definition event, description attribute are established according to the grammer of JavaMOP
With the several parts of self-defined user program.Monitor real time monitoring is ultimately produced, the step of monitor is generated according to JavaMOP tools
Suddenly model is generated into monitor, realizes on line real-time monitoring；
Step 2: Automatic parameter selection
1. the event recorder of monitor can from specification identification events definition, and institute in logging program control flow
There is the possibility entry of event.Event (being known as point of penetration in AspectJ) is used to match the tie point for meeting definition, including to certain
The calling of a method needs the domain changed, the expression formula to be calculated, the code etc. of the method just executed.Run time verification mould
Event definition part in type is the entrance of event recorder (Event Logger).
2. the event that the filter matching of monitor defines, but prevent every other event.
3. the parameter of capturing events is modified, and new parameter may return to the source finally continued.
A kind of Automatic parameter selection method of robot obstacle-avoiding of the present invention, advantage and effect are：By being tested in operation
Real-time adjust automatically parameter in model of a syndrome, the setting of the control system of active influence robot for danger zone, to
Verification model can be made preferably to adapt to the uncertain working environment of robot, the behavior of accurate measurements robot, avoid with
All static and dynamic barrier collisions.Here the parameter automatically selected refers to needing to combine actual environment situation indefinite
The parameter of Shi Genggai or technical staff can not determine the parameter of value in advance.Can be offline by Automatic parameter selection method,
Real-time undated parameter plays an important roll behavior restraint of mobile robot in the case of environment is uncertain.
Description of the drawings
Fig. 1 show the whole system frame diagram of arrangement run time verification environment of the embodiment of the present invention.
Fig. 2 show Automatic parameter selection flow chart of the embodiment of the present invention.
Fig. 3 show JavaMOP grammer sectional drawings and (comes from http://fsl.cs.illinois.edu/index.php/
JavaMOP4_Syntax)。
Specific implementation mode
With reference to the accompanying drawings and examples, the following further describes the technical solution of the present invention.
Hardware device, system framework, the Robot dodge strategy etc. that the embodiment of the present invention is based on first are introduced below.
(1) hardware device
The experiment porch of the embodiment of the present invention has selected EV3 robots, it is the third generation of Lego Company's exploitation
MINDSTORMS robots were listed in the second half year in 2013.It can load multiple sensors, such as ultrasonic sensor, touch
Sensor, gyroscope etc..Wherein, the embodiment of the present invention mainly uses ultrasonic sensor, touch sensor and two large-scale electricity
Machine (built-in angular sensor).
EV3 can not only load various sensors, but also can use leJOS, it is a kind of small-sized Java Virtual Machine,
Java language can be supported to program.LejOS EV3 supports run single java applet in robot, can also be long-range on PC
Remote-controlled robot.This pattern can be controlled in the EV3 in the leJOS EV3 menus in Eclipse environment controls option
System.LeJOS EV3 also have another pattern, it supports some programs to operate in PC, other programs operate in robot.
In addition, PC and robot can be by bluetooths, wireless or USB is mutually communicated.If wanting JavaMOP being attached to of the invention real
In the program for applying example, need to use latter mode, because the program directly run in EV3 robots cannot add Java generations
Reason.In addition, quickly can calculate and handle the data of EV3 transmission using the memory CPU of PC, then control command is returned to
Robot.This process is cycle, until program stopped.
In entire programming procedure, selected from " lejos.hardware.motor' " API document packets more complicated
Omniselector executes the basic operation of robot.In addition, the sample frequency of sensing data is set as 0.3s, ultrasonic sensor
Measurement accuracy be 1cm, measurement distance be 3cm to 250cm.
The actual working environment of robot is uncertain and complicated.Therefore, in experimental situation construction, static-obstacle is removed
Beyond the region of objective existence, also many dynamic, random barriers.They, which may be passerby or travelling vehicle, robot, will use ultrasonic wave
Sensor determines and avoiding obstacles.
(2) system framework
The embodiment of the present invention arranges run time verification environment, and whole system frame is as shown in Figure 1, include to carry multiple biographies
The robot system of sensor, tele-control system and run time verification part.
Run time verification tool dependency attribute, and attribute is to be described with logical language, but the language description of attribute is work
The major obstacle of Cheng Shi or user.Therefore, the first step, the demand that user or engineer propose must form turn to can be specific
The certain logic language of run time verification tool identification reads what related needs document manually completed by technical staff.Due to fortune
Monitoring must identify and capture the execution state of present procedure when row, thus the program information of tele-control system be also one must
The input of palpus.Run time verification part is added in tele-control system by way of adding monitoring agent for program,
It is reacted with the operation of program to monitor with the behavior of verifier people and violate the when of requiring in robot.
Tele-control system has usb data to connect and be wirelessly connected two kinds with the physical connection of robot system, leads to each other
Socket is crossed to transmit information.Tele-control system is the brain center of robot system, it determines the behavior side of robot
Formula.Content in conjunction with the sensing data from robot and the reaction from real-time verification are made order appropriate by it.
Then, robot system is sent a command to.Tele-control system includes the programmed logic of robot system, and robot system
Only it is responsible for simple behavior.Sensor that robot loads oneself or third-party, such as ultrasonic sensor, touch sensor
Deng.During robot motion, collect to cycle sensor in another thread data (for example, electric power, distance,
Increment of motion).It is communicated by socket with the tele-control system in LAN, and sensing data is transmitted to remotely
Control system simultaneously recycles the instruction for executing tele-control system.
(3) Robot dodge strategy
We devise three kinds of pre-contact security strategies to avoid obstacle：Strategy is turned to, strategy and emergency policy are retreated.Together
When, in order to reduce the damage of robot after collision by reducing speed and acceleration, increase security strategy after contact.
1, strategy is turned to：When detect front at a certain distance from have obstacle after, turn command is assigned to robot, then at
Continue after work(avoidance.
2, strategy is retreated：If encountering touch sensor or ultrasonic sensor measuring, machine too close from barrier
People just retreats when can not turn, avoiding obstacles of turning after retreating to certain distance.
3, emergency policy：Robot apart from obstacle closer to or touched, and can not complete retreat and turn behaviour
Make, wants terminator at this time.
4, security strategy after contacting：When can not avoiding obstacle, taking reduces the mode of speed and acceleration to reduce
Injury.
(4) Automatic parameter selection method
Parameter setting can be divided into preset parameter and dynamic parameter.The advantages of preset parameter is simple and convenient, but actually
Have some problems.The problem of enumerating two and encounter in an experiment：1) assume that risk distance is fixed as 0.3m, current robot with
The distance of barrier is 0.29m.Then in the ideal case, control terminal should send order to robot immediately.But due to
Physical presence delay (program execution time and wireless transmission time), robot will move on a distance during this period.2)
In another case, if current robot at a distance from barrier be 0.31m, control terminal can be identified as safety away from
From but this judgement has ignored the sampling time, that is to say, that have ignored robot before sensor device obtains data next time still
So in movement.This is abnormally dangerous, and perhaps robot encounters obstacle in this period.
Above-mentioned two problems highlight the shortcomings that preset parameter, they can not adapt to change.Compared with preset parameter, some
Special parameter would generally change with the variation of environment, or need the experience of engineer to determine, in order to more suitable
Environment is answered, safety and the robustness of system are improved.Meanwhile compared with the modification of program code itself, run time verification method
Field value or parameter can be changed offline in the program of operation, without terminator.Institute in this way especially suitable for
Emergency.
The dynamic process of parameter selection is described in detail in figure below, includes mainly following several steps, as shown in Figure 2：
1. event recorder can from specification identification events definition, and all events in logging program control flow
Possible entry.Event (being known as point of penetration in AspectJ) includes to some method for matching the tie point for meeting definition
It calls, needs the domain changed, the expression formula to be calculated, the code etc. of the method just executed.
2. the event that filter matching defines, but prevent every other event.
3. the parameter of capturing events is modified, and new parameter may return to the source finally continued.
In the example of Fig. 2, the event of definition is the calling to $ Function1Name $ methods.$ Parameter1 $ are used
A series of keyword captures, are then used as new parameter to return to source code.
Embodiment
(1) foundation and execution of run time verification model
Run time verification model (JavaMOP grammers sectional drawing such as Fig. 3 institutes of the model are established according to the grammer of JavaMOP
Show), including variable declarations, definition event, the several parts of description attribute and self-defined user program etc..Ultimately produce monitor reality
When the step of monitoring, monitor is generated according to JavaMOP tools model is generated into monitor, realize on line real-time monitoring,
It is as follows：
1. entering the file where mop files using dos command
This step of 2.javamop-d rvm-keepRVFiles*.mop generates .aj files and .rvm files
3.rv-monitor rvm/*.rvm rv-monitor orders generate monitoring library
4.javac rvm/monitor.java
Rm classes/mop/moniotr.java compilings monitor library file and remove java files
5.ajc-1.6-d weav rvm/*Aspect.aj rvm/*Monitor.java*.java weave file
6.javamopagent rvm/aspect.aj weav-n JavaMOPAgent-excludeJars
Establish the JavaMOP agencies of an entitled JavaMOPAgent
In the programmed environment of eclipse, by following steps：click"Run"tab->select"Run
configurations"tab->select the application you are running->select"Arguments"
tab->enter"javaagent:JavaMOPAgent.jar " into " VM options " textbox. are by monitor to act on behalf of
Form addition in a program, realize real time monitoring.
(2) Automatic parameter selection
Want dynamic arrange parameter to require to find the changing factor of threshold value S, in the case of robot obstacle-avoiding, threshold value S
Refer to the range of risk distance.When robot enters the range, need to take Robot dodge strategy appropriate in time.It was testing
Cheng Zhong, S are by current acceleration Acurr, present speed Vcurr sampling frequency F, the influence of delay time Tdelay, relationship
It is indicated with following formula：
D=Tdelay×Acurr+Vcurr+ε+W/2
S ∈ (W/2, D]
ε is that measurement error, Acurr and Vcurr are dynamically obtained in real time when robot is run, and W is the width of robot.
With the variation of speed and acceleration, ideal safe distance also changes therewith, is calculated most according to formula
Big safe distance and minimum safe distance difference are huge.The reason of minimum value occurs is that speed or acceleration are especially small
When, or perhaps when robot turning stopping or retrogressing stopping.The reason of maximum value occurs may be adding for moment super large
Speed, and it is also far apart from barrier at this time.Because of situation physical presence above, therefore we are to ideal safe distance
It has carried out constraint and has obtained actual safe distance.
1, when robot setting in motion, initial safe distance is set as default value D0；
2, when speed is 0, setting safe distance is default value D0；
3, when trolley meet barrier time be more than cycle time (delay time adds the period) when, safe distance is set as
Default value D0。
The default value D of safe distance herein0Computational methods it is as follows, wherein VaveIt is average speed：
D0=Tdelay×Vave+ε+W/2
Certainly, even if we have done above preparation to ensure collision prevention, but still there is the probability of very little that can occur
It collides, at this moment we contact security strategy to minimize impact force after being used as by the acceleration of limitation robot motion.It is testing
Robot travel speed is captured in model of a syndrome in real time, acceleration is calculated by speed data.When acceleration is more than limit value, swash
User-written subroutine living, provides warning and slows down.
In this invention, we devise suitable for load multiple sensors robot run time verification system
Frame.For robot obstacle-avoiding problem, reduce to injure from the aspect of two devising contact after avoiding obstacle and contact obstacle
Security strategy after preceding security strategy and contact.It is monitored and is handled by real-time online run time verification method.It is wherein dangerous
Distance parameter is updated according to the various uncertain factors such as delay time using Automatic parameter selection method dynamic.
Claims (1)
1. a kind of Automatic parameter selection method of robot obstacle-avoiding, is characterized in that：This method is as follows：
Step 1: the foundation and execution of run time verification model
Establish run time verification model according to the grammer of JavaMOP, including variable declarations, define event, description attribute and from
Define the several parts of user program；The step of ultimately producing monitor real time monitoring, monitor is generated according to JavaMOP tools will
Model generates monitor, realizes on line real-time monitoring；
Step 2: Automatic parameter selection
(1) definition of the event recorder of monitor identification events from specification, and all events in logging program control flow
Possibility entry；Event includes the calling to some method, needs the domain changed, for matching the tie point for meeting definition
The expression formula of calculating, the code of the method just executed；Event definition part in run time verification model is event recorder
Entrance；
(2) event that the filter matching of monitor defines, but prevent every other event；
(3) parameter of capturing events is modified, and new parameter may return to the source finally continued.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810468201.2A CN108681447B (en) | 2018-05-16 | 2018-05-16 | Automatic parameter selection method for robot obstacle avoidance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810468201.2A CN108681447B (en) | 2018-05-16 | 2018-05-16 | Automatic parameter selection method for robot obstacle avoidance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108681447A true CN108681447A (en) | 2018-10-19 |
| CN108681447B CN108681447B (en) | 2021-09-17 |
Family
ID=63805551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810468201.2A Active CN108681447B (en) | 2018-05-16 | 2018-05-16 | Automatic parameter selection method for robot obstacle avoidance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108681447B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113885520A (en) * | 2021-10-27 | 2022-01-04 | 南京理工大学 | Method for determining dynamic safety distance between autonomous mobile machine and human body |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5908458A (en) * | 1997-02-06 | 1999-06-01 | Carnegie Mellon Technical Transfer | Automated system and method for control of movement using parameterized scripts |
| CN104020674A (en) * | 2014-04-21 | 2014-09-03 | 华南农业大学 | Matlab simulation visualized platform of vehicle Bug obstacle avoidance algorithm |
| CN106363600A (en) * | 2016-10-18 | 2017-02-01 | 天津大学 | Self-propelled intelligent robot system for indoor environment measurement |
| CN106444790A (en) * | 2016-12-09 | 2017-02-22 | 屈兆辉 | Industrial AGV system based on PLC control |
| US20170142156A1 (en) * | 2015-11-12 | 2017-05-18 | Toyota Infotechnology Center Usa, Inc. | Application Assurance for Open Platform In-Vehicle Infotainment System |
| CN106940562A (en) * | 2017-03-09 | 2017-07-11 | 华南理工大学 | A kind of mobile robot wireless clustered system and neutral net vision navigation method |
-
2018
- 2018-05-16 CN CN201810468201.2A patent/CN108681447B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5908458A (en) * | 1997-02-06 | 1999-06-01 | Carnegie Mellon Technical Transfer | Automated system and method for control of movement using parameterized scripts |
| CN104020674A (en) * | 2014-04-21 | 2014-09-03 | 华南农业大学 | Matlab simulation visualized platform of vehicle Bug obstacle avoidance algorithm |
| US20170142156A1 (en) * | 2015-11-12 | 2017-05-18 | Toyota Infotechnology Center Usa, Inc. | Application Assurance for Open Platform In-Vehicle Infotainment System |
| CN106363600A (en) * | 2016-10-18 | 2017-02-01 | 天津大学 | Self-propelled intelligent robot system for indoor environment measurement |
| CN106444790A (en) * | 2016-12-09 | 2017-02-22 | 屈兆辉 | Industrial AGV system based on PLC control |
| CN106940562A (en) * | 2017-03-09 | 2017-07-11 | 华南理工大学 | A kind of mobile robot wireless clustered system and neutral net vision navigation method |
Non-Patent Citations (6)
| Title |
|---|
| ANKUSH DESAI: "DRONA: A Framework for Safe Distributed Mobile Robotics", 《2017 ACM/IEEE 8TH INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS (ICCPS)》 * |
| KANG YU: "A Predictive Runtime Verification Framework for Cyber-Physical Systems", 《2014 IEEE EIGHTH INTERNATIONAL CONFERENCE ON SOFTWARE SECURITY AND RELIABILITY-COMPANION》 * |
| PATRICK O’NEIL MEREDITH: "An overview of the MOP runtime verification framework", 《 INTERNATIONAL JOURNAL ON SOFTWARE TOOLS FOR TECHNOLOGY TRANSFER》 * |
| RUI WANG: "From Offline Towards Real-Time Verification for Robot Systems", 《 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS》 * |
| 周戈: "基于运行时验证的监控器生成技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王珍: "参数化运行时监控研究", 《计算机科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113885520A (en) * | 2021-10-27 | 2022-01-04 | 南京理工大学 | Method for determining dynamic safety distance between autonomous mobile machine and human body |
| CN113885520B (en) * | 2021-10-27 | 2022-04-22 | 南京理工大学 | Method for determining dynamic safety distance between autonomous mobile machine and human body |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108681447B (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105058396A (en) | Robot teaching system and control method thereof | |
| CN108828944B (en) | Encoder fault diagnosis system and method based on improved PSO and SVM | |
| KR102025100B1 (en) | System and method for monitoring robot motion | |
| Popov et al. | Collision detection, localization & classification for industrial robots with joint torque sensors | |
| US9982414B2 (en) | Operation identification of a work machine | |
| CN109213074A (en) | control device and machine learning device | |
| CN102426391A (en) | Method for determining whether there is collision during robot operation | |
| KR20180013757A (en) | Using human motion sensors to detect movement when in the vicinity of hydraulic robots | |
| CN107116553B (en) | Mechanical arm operation method and device | |
| JP7243979B2 (en) | Robot interference determination device, robot interference determination method, robot control device, robot control system, human motion prediction device, and human motion prediction method | |
| Millard et al. | Run-time detection of faults in autonomous mobile robots based on the comparison of simulated and real robot behaviour | |
| CN109079800A (en) | Robot collision checking method, device, storage medium, processor and robot | |
| CN108681447A (en) | A kind of Automatic parameter selection method of robot obstacle-avoiding | |
| Cheng et al. | Harnessing evolutionary computation to enable dynamically adaptive systems to manage uncertainty | |
| Årzén et al. | A component-based approach to the design of networked control systems | |
| Luo et al. | Runtime verification of robots collision avoidance case study | |
| JP2023547612A (en) | Safety systems and methods used in robot operation | |
| CN110914019B (en) | Robot stopping distance simulation method | |
| CN111331603A (en) | Stress type motion posture conversion method and system for wheel-legged robot | |
| CN110549375A (en) | protective door anti-collision method and system for mechanical arm | |
| US20230069482A1 (en) | Machine learning-based environment fail-safes through multiple camera views | |
| EP4000815A1 (en) | Visualization of a protected area | |
| US11318606B2 (en) | Controlling an automation assembly | |
| Vassev et al. | Prototyping home automation wireless sensor networks with ASSL | |
| Zhang et al. | Formally verifying navigation safety for ground robots |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |