JP2017009572A5

JP2017009572A5 -

Info

Publication number: JP2017009572A5
Application number: JP2015198675A
Authority: JP
Filing date: 2015-10-06
Publication date: 2018-07-12
Anticipated expiration: 2035-10-06

Description

本開示の一態様に係る車載レーダ装置は、検知範囲に対してレーダ信号をフレーム毎に送信し、前記レーダ信号が１つ以上の物体に反射された１つ以上の反射信号を受信する送受信部と、前記検知範囲内の各方位において、前記１つ以上の反射信号に基づいて検知される１つ以上の反射点のうち、車載レーダ装置からの距離が最も近い反射点の位置を、前記検知範囲内において前記１つ以上の物体が存在しない領域の境界候補位置として、前記フレーム毎に検出する検出部と、前記車載レーダ装置の移動量に関する移動量データを計算する計算部と、前記移動量データに基づいて、過去のフレームにおいて検出された前記境界候補位置を現フレームでの境界位置に変換することによって、推定された推定境界位置を生成する推定部と、前記現フレームの前記境界候補位置と前記推定境界位置とを用いてスムージング処理し、前記検知範囲内において前記１つ以上の物体が存在しない領域との境界位置の有無を判断し、前記境界位置を有すると判断した場合、前記境界位置を運転支援装置に出力するスムージング部と、を具備する構成を採る。 A vehicle-mounted radar device according to one aspect of the present disclosure transmits / receives a radar signal to a detection range for each frame, and receives one or more reflected signals reflected by one or more objects. And, in each direction within the detection range, among the one or more reflection points detected based on the one or more reflection signals, the position of the reflection point closest to the on-vehicle radar device is detected. As a boundary candidate position of a region where the one or more objects do not exist within a range, a detection unit that detects each frame, a calculation unit that calculates movement amount data related to the movement amount of the in-vehicle radar device, and the movement amount An estimation unit that generates an estimated estimated boundary position by converting the boundary candidate position detected in a past frame into a boundary position in a current frame based on the data; and And smoothing processing using said boundary candidate position of frame and the said estimated boundary position, it is determined whether there is a boundary position between the one or more objects does not exist a region within the detection range and having a boundary position When it is determined, a configuration including a smoothing unit that outputs the boundary position to the driving support device is adopted.

本開示の一態様に係る領域検出方法は、検知範囲に対してレーダ信号をフレーム毎に送信し、前記レーダ信号が１つ以上の物体に反射された１つ以上の反射信号を受信し、前記検知範囲内の各方位において、前記１つ以上の反射信号に基づいて検知される１つ以上の反射点のうち、車載レーダ装置からの距離が最も近い反射点の位置を、前記検知範囲内において前記１つ以上の物体が存在しない領域の境界候補位置として、フレーム毎に検出し、前記車載レーダ装置の移動量に関する移動量データを計算し、前記移動量データに基づいて、過去のフレームにおいて検出された前記境界候補位置を現フレームでの境界位置に変換することによって、推定された推定境界位置を生成し、前記現フレームの前記境界候補位置と前記推定境界位置とを用いてスムージング処理し、前記検知範囲内において物体が存在しない領域との境界位置の有無を判断し、前記境界位置を有すると判断した場合、前記境界位置を運転支援装置に出力する。 In the region detection method according to an aspect of the present disclosure, a radar signal is transmitted to a detection range for each frame, and the radar signal receives one or more reflected signals reflected by one or more objects, Among each of the one or more reflection points detected based on the one or more reflection signals in each azimuth within the detection range, the position of the reflection point closest to the in-vehicle radar device is determined within the detection range. Detected for each frame as a boundary candidate position of the area where the one or more objects do not exist, calculates movement amount data related to the movement amount of the in-vehicle radar device, and detects in a past frame based on the movement amount data The estimated boundary position is generated by converting the determined boundary candidate position into a boundary position in the current frame, and the boundary candidate position and the estimated boundary position in the current frame are There was smoothed, to determine whether the boundary position between the region where no object exists within the detection range, if it is determined to have the boundary position, and outputs the boundary position in the driving support device.

Claims

A transmission / reception unit that transmits a radar signal to the detection range for each frame and receives one or more reflected signals of the radar signal reflected by one or more objects;
In each azimuth within the detection range, among the one or more reflection points detected based on the one or more reflection signals, the position of the reflection point closest to the in-vehicle radar device is determined within the detection range. A detection unit that detects for each frame as a boundary candidate position of a region where the one or more objects do not exist;
A calculation unit for calculating movement amount data related to the movement amount of the in-vehicle radar device;
An estimation unit that generates an estimated estimated boundary position by converting the boundary candidate position detected in a past frame into a boundary position in a current frame based on the movement amount data;
Smoothing processing is performed using the boundary candidate position and the estimated boundary position of the current frame, the presence / absence of a boundary position with the region where the one or more objects do not exist in the detection range, and the boundary position A smoothing unit that outputs the boundary position to the driving support device ,
An on-vehicle radar device comprising:

The smoothing unit is
When a likelihood value is calculated using a preset likelihood weight for the boundary candidate position or the estimated boundary position in each direction within the detection range, and the likelihood value is equal to or greater than a predetermined threshold value, It determines that the boundary position exists,
When the boundary position in each direction in the detection range is present, the weighted distance of said boundary candidate position or the estimated boundary position of the current frame by using the output value weight set in advance, using the weighted average Output the calculated output distance as the boundary position,
The on-vehicle radar device according to claim 1.

The one or more objects include stationary objects and moving objects;
The on-vehicle radar device according to claim 1.

The one or more objects are stationary objects;
The on-vehicle radar device according to claim 1.

A camera unit for photographing the detection range;
An area setting unit that sets an image area used for object recognition in the captured video, based on the boundary position;
An object recognition unit that performs object recognition on the set image region;
Further including
The on-vehicle radar device according to claim 1.

Transmitting a radar signal to the detection range for each frame, and receiving one or more reflected signals in which the radar signal is reflected by one or more objects;
In each azimuth within the detection range, among the one or more reflection points detected based on the one or more reflection signals, the position of the reflection point closest to the in-vehicle radar device is determined within the detection range. , Detecting for each frame as a boundary candidate position of the region where the one or more objects do not exist,
Calculate movement amount data relating to the movement amount of the in-vehicle radar device,
Based on the movement amount data, the estimated boundary position is generated by converting the boundary candidate position detected in the past frame into the boundary position in the current frame,
When smoothing processing is performed using the boundary candidate position and the estimated boundary position of the current frame, it is determined whether or not there is a boundary position with an area where no object exists in the detection range, and the boundary position is determined , Outputting the boundary position to the driving support device ,
Region detection method.

The smoothing process includes
When a likelihood value is calculated using a preset likelihood weight for the boundary candidate position or the estimated boundary position in each direction within the detection range, and the likelihood value is equal to or greater than a predetermined threshold value, Determining that the boundary position exists;
When the boundary position exists in each direction within the detection range, the weight of the boundary candidate position or the estimated boundary position of the current frame is weighted using a preset output value weight, and a weighted average is used. Output the calculated output distance as the boundary position,
The region detection method according to claim 6.

The one or more objects include stationary objects and moving objects;
The region detection method according to claim 6.

The one or more objects are stationary objects;
The region detection method according to claim 6.

The detection range is photographed using a camera unit,
An image region used for object recognition among the captured images is set based on the boundary position,
Object recognition is performed on the set image area.
Further including
The region detection method according to claim 6.