Determining Ego Motion
A computer implemented method to determine ego motion of a vehicle, the vehicle having at least one radar emitter with a plurality of reception antennae, the method including the operations of acquiring, from the reception antennae, different frames of radar data of the vehicle surrounding environment, each frame being acquired at a different time; deriving from the radar data of each different frame, an environment map of the vehicle surrounding environment; and deriving the ego motion of the vehicle by: merging environment maps from at least two different frames into one accumulated map, computing, from the accumulated map, a motion vector for each pixel of the accumulated map, and extracting, from the accumulated map, a mask map including a tensor mapping a weight for each pixel of the accumulated map.
1 . A computer implemented method to determine ego motion of a vehicle having at least one radar emitter with a plurality of reception antennae, the method comprising:
acquiring, from the reception antennae, different frames of radar data of a vehicle surrounding environment, each frame being acquired at a different time;
deriving from the radar data of each different frame, an environment map of the vehicle surrounding environment; and
deriving the ego motion of the vehicle by:
merging environment maps from at least two different frames into one accumulated map;
computing, from the accumulated map, a motion vector for each pixel of the accumulated map; and
extracting, from the accumulated map, a mask map including a tensor mapping a weight for each pixel of the accumulated map.
2 . The computer implemented method of claim 1 , wherein deriving the ego motion includes:
associating each motion vector with a weight of the mask map.
3 . The computer implemented method of claim 1 , wherein deriving the ego motion includes:
computing a translation and a rotation of the vehicle surrounding environment introduced by a movement of the vehicle, based on the computed motion vectors and the extracted mask map.
4 . The computer implemented method of claim 1 , wherein deriving the ego motion includes:
deriving a x-speed, a y-speed, and a yaw speed of the vehicle.
5 . The computer implemented method of claim 1 , further including:
extracting reference points from the mask map, the reference points being associated with objects of the surrounding environment of the vehicle.
6 . The computer implemented method of claim 1 , further including:
graphically representing at least one of the mask map or the environment map to a display of the vehicle.
7 . The computer implemented method of claim 1 , further including:
graphically representing one or more characteristics extracted from at least one of the mask map or the environment map.
8 . The computer implemented method of claim 1 , wherein computing a motion vector for each pixel of the accumulated map and extracting the mask map is performed by one or more layers of a trained neural network.
9 . The computer implemented method of claim 1 , wherein the radar data includes a plurality of Doppler channels and deriving an environment map of the vehicle surrounding environment includes:
reducing the plurality of Doppler channels to one feature channel.
10 . The computer implemented method of claim 1 , further including:
controlling one or more functions of an advanced driver-assistance system (ADAS) of a vehicle.
11 . A method for training a neural network for determining ego motion of a vehicle, the method comprising:
providing different frames of radar data of a training vehicle surrounding environment and at least one of corresponding localization or driving information of the training vehicle;
deriving ego motion from the radar data by:
acquiring different frames of radar data of the training vehicle surrounding environment, each frame being acquired at a different time;
deriving from the radar data of each different frame, an environment map of the vehicle surrounding environment; and
deriving the ego motion of the vehicle by:
merging environment maps from at least two different frames into one accumulated map;
computing, from the accumulated map, a motion vector for each pixel of the accumulated map; and
extracting, from the accumulated map, a mask map including a tensor mapping a weight for each pixel of the accumulated map;
deriving ground truth ego motion based on the at least one of localization or driving information; and
determining optimal weights of a neural network by minimizing a loss function between the ego motion as an input and the ground truth ego motion as a target output.
12 . The method of claim 11 , wherein deriving the ego motion includes:
associating each motion vector with a weight of the mask map.
13 . The method of claim 11 , wherein deriving the ego motion includes:
computing a translation and a rotation of the vehicle surrounding environment introduced by a movement of the vehicle, based on the computed motion vectors and the extracted mask map.
14 . The method of claim 11 , wherein deriving the ego motion includes:
deriving a x-speed, a y-speed, and a yaw speed of the vehicle.
15 . The method of claim 11 , further including:
extracting reference points from the mask map, the reference points being associated with objects of the surrounding environment of the vehicle.
16 . The method of claim 11 , further including at least one of:
graphically representing at least one of the mask map or the environment map to a display of the vehicle; or
graphically representing one or more characteristics extracted from at least one of the mask map or the environment map.
17 . The method of claim 11 , wherein computing a motion vector for each pixel of the accumulated map and extracting the mask map is performed by one or more layers of the trained neural network.
18 . The method of claim 11 , wherein the radar data includes a plurality of Doppler channels and deriving an environment map of the vehicle surrounding environment includes:
reducing the plurality of Doppler channels to one feature channel.
19 . The method of claim 11 , further including:
controlling one or more functions of an advanced driver-assistance system (ADAS) of a vehicle.
20 . A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to:
acquire, from a vehicle radar emitter with a plurality of reception antennae, different frames of radar data of a vehicle surrounding environment, each frame being acquired at a different time;
derive, from the radar data of each different frame, an environment map of the vehicle surrounding environment; and
derive ego motion of the vehicle by:
merging environment maps from at least two different frames into one accumulated map;
computing, from the accumulated map, a motion vector for each pixel of the accumulated map; and
extracting, from the accumulated map, a mask map including a tensor mapping a weight for each pixel of the accumulated map.