Font Size: a A A

Research Of Software For Mobile Robot Navigation And Partial Key Technology

Posted on:2015-03-10Degree:MasterType:Thesis
Country:ChinaCandidate:L B GuanFull Text:PDF
GTID:2268330425981430Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
With the development of power electronics technology, computer technology, image processing, machine intelligence and other related technologies, the performance of robot continues to be improved, and it has gradually been widely used in industry, agriculture, health care, services and other fields. As mobile robot is often in motion, it has higher demands on the robot than ordinary robot, such as more flexible movement mechanism, stronger external perception and planning capacity of local autonomy. Based on this kind of demand, mobile robot navigation technology has gradually been developed. In certain applications, mobile robot navigation technology has already been used.On the basis of the features of inertial navigation for mobile robot in small area, this paper first designs and implements navigation software functions for upper computer, i.e. PC, including map matching, operations and human-machine interface, and in the lower computer, i.e. ARM mcu, implements an improved dead reckoning algorithm, and also works out communication protocols which serve the robot both for normal navigation work and debugging. Completed work includes the following three parts:(1) Implementation of PC navigation software based on MFC, including a simple and practical method of making the target area map. Implementation of efficient serial communication program, with serial API communication functions, combined with non-blocking communication, multi-threading and other means. Implementation of a series of commonly used functions for maps with interface provided by MapX, such as zoom in, zoom out, pan, layer management.(2)Software design for lower computer based on RL-RTX real-time operating system. An improved dead reckoning algorithm is proposed, aiming at the drawbacks of the traditional dead reckoning algorithm, thus the robot navigation can adapt to non-structural road environment. Meanwhile, the code is implemented in the framework of RL-RTX system. Implementation of a complete data communication protocol, which allows the data transmission both when the robot is working and debugging, and the communication framework is also designed.(3) On the basis of the features of non-GPS mobile robot navigation, a scheme of map matching for real area and the navigation maps is proposed, which needs only two points as train points. Currently, non-GPS navigation systems (mostly inertial navigation systems) require the original map providing precise north-south orientation and accurate map scale, while the demand for maps of the proposed scheme is very low, only requiring that the ratio of length and width of the map itself is same to that of the target area. Besides, the map process is also very simple. Experimental results show that the solution is accurate and efficient. Meanwhile, againt to the problem of inevitable error in inertial navigation, this paper designs an artificial means to compensate for the navigation error correction.In summary, this paper provides the detailed design and implementation for PC navigation software and lower computer software, and proposes a practical map-matching scheme, thus providing a new and effective solution for the field of robot navigation.
Keywords/Search Tags:Mobile robot, Inertial navigation, Area navigation, Map matching, Dead reckoning
PDF Full Text Request
Related items