Search

CN-122017907-A - Error real-time compensation method for GNSS dynamic positioning

CN122017907ACN 122017907 ACN122017907 ACN 122017907ACN-122017907-A

Abstract

The invention discloses an error real-time compensation method for GNSS dynamic positioning, which relates to the field of GNSS dynamic positioning and comprises the steps of obtaining track segments to be corrected with continuous motion trend, retrieving digital map data of corresponding areas from a local storage and/or cloud servers, determining one or more candidate matching roads according to the matching degree of track heading change characteristics and road curvature characteristics, conducting short-time dead reckoning to obtain a predicted position at the current moment, verifying projection points based on road passing direction constraint and connectivity constraint, and taking a final corrected positioning result as a datum point of dead reckoning at the next moment to continuously and progressively correct the GNSS dynamic positioning error. The method has the advantages that the dynamic positioning error and the inertial accumulated drift are restrained through the short-time dead reckoning prediction, the forced positioning result is always and reasonably attached to a real road network, and the high-reliability, low-cost, continuous and stable dynamic positioning is realized.

Inventors

  • LIN JIE

Assignees

  • 深圳市瑞图同创科技有限公司

Dates

Publication Date
20260512
Application Date
20260313

Claims (9)

  1. 1. A method for compensating errors in real time for dynamic positioning of a GNSS, comprising: acquiring an original positioning sequence output by a GNSS receiver in real time, and performing smooth filtering treatment on the original positioning sequence to obtain a track segment to be corrected with a continuous motion trend; according to the position range of the track segment to be corrected, digital map data of the corresponding region are called from a local storage and/or cloud server; comparing the similarity of the track segment to be corrected with candidate roads in the digital map, and determining one or more candidate matching roads according to the matching degree of the track course change characteristics and the road curvature characteristics; Carrying out short-time dead reckoning by utilizing the carrier movement speed and direction information output by the inertial measurement unit and combining the final correction positioning result output at the last moment to obtain the predicted position at the current moment; projecting the predicted position onto a candidate matching road, checking the projection point based on the road passing direction constraint and connectivity constraint, and setting the projection point after the checking to be a final corrected positioning result; and taking the final correction positioning result as a reference point of dead reckoning at the next moment, updating the track segment to be corrected in real time, and carrying out road matching and topology constraint correction again on the updated track segment to be corrected to realize continuous progressive correction of the GNSS dynamic positioning error.
  2. 2. The method for compensating errors in dynamic positioning of GNSS according to claim 1, wherein the obtaining, in real time, an original positioning sequence output by a GNSS receiver, and performing smoothing filtering processing on the original positioning sequence, and obtaining a track segment to be corrected with a continuous motion trend specifically includes: Acquiring an original positioning sequence output by a GNSS receiver, and removing isolated noise points in the original positioning sequence based on kinematic constraint; Carrying out Kalman filtering smoothing treatment on the positioning sequence from which the noise points are removed; and intercepting a continuous track with a preset length from the smoothed positioning sequence as a track segment to be corrected.
  3. 3. The method for compensating errors in GNSS dynamic positioning according to claim 2, wherein the retrieving digital map data of the corresponding area from the local storage and/or cloud server according to the position range of the track segment to be corrected specifically includes: acquiring longitude and latitude coordinates of all positioning points in the track segment to be corrected, and calculating an outsourcing rectangular range of the track segment to be corrected as a calling area of the digital map; According to the calling area, preferentially retrieving the digital map data of the corresponding area from the local storage, and if the complete map data covering the calling area exists in the local storage, directly loading and using; if the complete map data covering the access area does not exist in the local storage, a map access request is sent to the cloud server, and the digital map data of the corresponding area is downloaded and cached to the local storage.
  4. 4. The method for compensating errors in GNSS dynamic positioning according to claim 3, wherein the comparing the track segments to be corrected with the candidate roads in the digital map according to the matching degree of the track course change feature and the road curvature feature, determining one or more candidate matching roads specifically includes: Extracting the course angle of each positioning point in the track segment to be corrected, and constructing a track course angle change sequence; extracting curvature change characteristics of each candidate road center line in the digital map, and constructing a road curvature characteristic sequence; Calculating the similarity distance between the track course angle change sequence and each candidate road curvature characteristic sequence by adopting a dynamic time warping algorithm; And selecting a plurality of roads with minimum similarity distance as candidate matching roads.
  5. 5. The method for compensating errors in dynamic positioning of GNSS of claim 4, wherein said obtaining a predicted position at a current time by using information on a carrier movement speed and a direction output by the inertial measurement unit and combining a final corrected positioning result output at a previous time to perform short-time dead reckoning specifically comprises: acquiring triaxial acceleration and angular velocity data output by an inertial measurement unit, and calculating the instantaneous motion speed and heading angle of the carrier through a mechanical programming algorithm; The final corrected positioning result output at the previous moment is taken as a starting datum point, and the displacement increment in the time interval from the previous moment to the current moment is calculated by combining the instantaneous movement speed and the heading angle; and accumulating the displacement increment to the final corrected positioning result at the previous moment to obtain the predicted position coordinate at the current moment.
  6. 6. The method for compensating errors in GNSS dynamic positioning according to claim 5, wherein projecting the predicted position onto the candidate matching road, verifying the projected point based on the road traffic direction constraint and the connectivity constraint, and setting the projected point after the verification as the final corrected positioning result specifically includes: for each candidate matching road, vertically projecting the predicted position at the current moment onto the central line of the road to obtain corresponding projection point coordinates; calculating an included angle between the motion direction of the predicted position and the extending direction of the candidate matching road at the projection point; judging whether the included angle is smaller than a preset direction consistency threshold value, if so, passing direction constraint checking of the projection point is qualified, taking the projection point as a candidate point, and entering a connectivity constraint checking link; Judging whether a topological communication relation exists between the current candidate matching road and the road where the final corrected positioning result is located at the previous moment, and judging whether the communication path distance is matched with the estimated driving distance or not; And regarding the projection points passing the connectivity constraint verification as effective projection points, if a plurality of effective projection points exist, carrying out weighted summation scoring according to the projection distance, the direction consistency and the connectivity, and selecting the projection point with the highest scoring as the final corrected positioning result at the current moment.
  7. 7. The method for compensating errors in dynamic positioning of GNSS of claim 6, wherein said continuously progressive correction of errors in dynamic positioning of GNSS comprises: Storing the final corrected positioning result output at the current moment into a history track buffer memory, and taking the final corrected positioning result as a starting datum point of dead reckoning at the next moment; Adding a final correction positioning result at the current moment into the track segment to be corrected, removing a positioning point at the earliest moment in the track segment to be corrected, keeping the length of the track segment to be corrected constant, and obtaining an updated track segment to be corrected; Re-executing the road matching and topology constraint correction flow based on the updated track segment to be corrected to obtain a final correction positioning result at the next moment; and (3) in the cyclic updating and correcting process, the positioning result at each moment is subjected to iterative optimization by combining with the latest map constraint on the basis of the correction at the previous moment, so that the continuous progressive correction of the GNSS dynamic positioning error is realized.
  8. 8. An electronic device comprising at least one processor, and a memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method for error real-time compensation for GNSS dynamic positioning according to any of claims 1-7.
  9. 9. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements a method for error real-time compensation for GNSS dynamic positioning according to any of the claims 1-8.

Description

Error real-time compensation method for GNSS dynamic positioning Technical Field The invention relates to the field of GNSS dynamic positioning, in particular to a real-time error compensation method for GNSS dynamic positioning. Background With the wide application of Global Navigation Satellite Systems (GNSS) in the fields of intelligent transportation, automatic driving, mobile mapping and the like, high-precision and high-reliability positioning in a dynamic environment has become a core foundation for guaranteeing the safe operation of related systems. However, in complex scenes such as urban canyons, under overhead bridges, tunnel entrances and exits, GNSS signals are easily interfered by multipath effects, non-line-of-sight reception, satellite signal shielding and other factors, so that problems such as rough jump, signal interruption or continuity interruption occur in an original positioning sequence, and the usability and user experience of positioning results are seriously affected. To address this challenge, inertial Measurement Units (IMUs) and GNSS are commonly used in the industry to perform integrated navigation, and short-time high-precision relative measurement capabilities of the IMUs are utilized to make up for the shortfall of GNSS. However, the problem of logic deviation between the positioning track and the real road network is still difficult to solve by relying on the loose combination or the tight combination scheme of the GNSS/IMU alone, namely, the positioning point is smooth in the mathematical statistics sense, but may appear in the interior of a building, a facing lane or a no-road area, so that navigation guidance is invalid or the driving assistance system is misjudged. It should be further noted that, a great deal of research in the current academia and industry focuses on improving the absolute accuracy of GNSS positioning, and attempts to compress dynamic positioning errors to the level of decimeters or even centimeters by means of real-time kinematic (RTK), precision single point positioning (PPP), satellite Based Augmentation System (SBAS) and other technical means. However, in a real urban dynamic environment, continuous and stable centimeter-level absolute positioning is difficult to achieve, on one hand, RTK and other technologies are highly dependent on the coverage density and data link reliability of a reference station network, fixed unlocking is easy to occur in a signal shielding area, so that accuracy is rapidly degraded, on the other hand, centimeter-level accuracy maintenance needs high-quality satellite observation values and complex atmospheric error modeling, and on the other hand, problems of intermittent signals, frequent cycle slip and the like in urban canyons make high-accuracy ambiguity fixing extremely difficult. Even if centimeter-level positioning is realized at certain moments, if the positioning point falls out of the road range by mistake due to multipath effect, the "high precision" becomes a hidden danger of misleading the system, because it is more practical meaning for navigation decision that "whether on the right road" is far more accurate than "how accurate the absolute coordinates are". Therefore, in the process of pursuing absolute precision, the logic rationality and scene adaptability of the positioning result are often ignored in the prior art. In a complex urban environment, the limitation of signal physical characteristics is that the precision improvement which depends on GNSS observation values only finally encounters a bottleneck, and how to utilize low-cost sensors and priori geographic information to realize continuous error suppression and correction on the premise that the positioning result is always attached to a road network becomes a key technical problem to be solved urgently in the dynamic positioning field. Disclosure of Invention In order to solve the technical problems, the technical scheme provides an error real-time compensation method for GNSS dynamic positioning, which solves the problems that in the complex urban environment, due to the limitation of signal physical characteristics, the precision improvement which is simply dependent on GNSS observation values finally encounters a bottleneck, and how to utilize low-cost sensors and priori geographic information, and realize continuous suppression and correction of errors on the premise that the positioning result is always attached to a road network. In order to achieve the above purpose, the invention adopts the following technical scheme: an error real-time compensation method for GNSS dynamic positioning, comprising: acquiring an original positioning sequence output by a GNSS receiver in real time, and performing smooth filtering treatment on the original positioning sequence to obtain a track segment to be corrected with a continuous motion trend; according to the position range of the track segment to be corrected, digital map data of the corresponding