Systems and methods for estimating cuboids from LiDAR, map and image data

A method for operating an autonomous vehicle, comprising:

• obtaining, by a computing device, a LiDAR dataset;
• plotting, by a computing device, the LiDAR dataset on a 3D graph to define a 3D point cloud;
• using, by a computing device, the LiDAR dataset and contents of a vector map to define a cuboid on the 3D graph that encompasses points of the 3D point cloud that are associated with an object in proximity to the vehicle, where the vector map comprises lane information;
• and
• using the cuboid to facilitate driving-related operations of the autonomous vehicle;
• wherein the cuboid is defined on the 3D graph by inferring a heading distribution Ppt(θ) for the object from the 3D point cloud;
• wherein the heading distribution Ppt(θ) is defined by mathematical equation Ppt(θ)=1/η*exp(−C(θ)), where exp( ) is the exponential function, η is a normalization constant, and C(θ) is a cost function.

A method for operating an autonomous vehicle, comprising:

• obtaining, by a computing device, a LiDAR dataset;
• plotting, by a computing device, the LiDAR dataset on a 3D graph to define a 3D point cloud;
• using, by a computing device, the LiDAR dataset and contents of a vector map to define a cuboid on the 3D graph that encompasses points of the 3D point cloud that are associated with an object in proximity to the vehicle, where the vector map comprises lane information;
• and
• using the cuboid to facilitate driving-related operations of the autonomous vehicle;
• wherein the cuboid is defined on the 3D graph by determining a heading distribution Pm(θ) from the vector map;
• and
• wherein the heading distribution Pm(θ) is defined by mathematical equation pm(θ)=maxi(N(θ:
• αi, σi2), where N( ) is a normal distribution, αi are the lane directions, and σi2 are the corresponding variances which are configurable.

A system, comprising:

• a processor;
• a non-transitory computer-readable storage medium comprising programming instructions that are configured to cause the processor to implement a method for operating a vehicle, wherein the programming instructions comprise instructions to:
• obtain a LiDAR dataset;
• plot the LiDAR dataset on a 3D graph to define a 3D point cloud;
• use the LiDAR dataset and contents of a vector map to define a cuboid on the 3D graph that encompasses points of the 3D point cloud that are associated with an object in proximity to the vehicle, where the vector map comprises lane information;
• and
• use the cuboid to facilitate driving-related operations of the autonomous vehicle;
• wherein the cuboid is defined on the 3D graph by inferring a heading distribution Ppt(θ) for the object from the 3D point cloud, determining a heading distribution Pm(θ) from the vector map, and performing filter operations to obtain a posterior distribution of a heading using the heading distributions Ppt(θ) and Pm(θ).

A system, comprising:

• a processor;
• a non-transitory computer-readable storage medium comprising programming instructions that are configured to cause the processor to implement a method for operating a vehicle, wherein the programming instructions comprise instructions to:
• obtain a LiDAR dataset;
• plot the LiDAR dataset on a 3D graph to define a 3D point cloud;
• use the LiDAR dataset and contents of a vector map to define a cuboid on the 3D graph that encompasses points of the 3D point cloud that are associated with an object in proximity to the vehicle, where the vector map comprises lane information;
• and
• use the cuboid to facilitate driving-related operations of the autonomous vehicle;
• wherein the cuboid is defined on the 3D graph by inferring a heading distribution Ppt(θ) for the object from the 3D point cloud;
• and
• wherein the heading distribution Ppt(θ) is defined by mathematical equation Ppt(θ)=1/η*exp(−C(θ)), where exp( ) is the exponential function, η is a normalization constant, and C(θ) is a cost function.

A system, comprising:

• a processor;
• a non-transitory computer-readable storage medium comprising programming instructions that are configured to cause the processor to implement a method for operating a vehicle, wherein the programming instructions comprise instructions to:
• obtain a LiDAR dataset;
• plot the LiDAR dataset on a 3D graph to define a 3D point cloud;
• use the LiDAR dataset and contents of a vector map to define a cuboid on the 3D graph that encompasses points of the 3D point cloud that are associated with an object in proximity to the vehicle, where the vector map comprises lane information;
• and
• use the cuboid to facilitate driving-related operations of the autonomous vehicle;
• wherein the cuboid is defined on the 3D graph by determining a heading distribution Pm(θ) from the vector map;
• and
• wherein the heading distribution Pm(θ) is defined by mathematical equation pm(θ)=maxi(N(θ:
• αi, σi2), where N( ) is a normal distribution, αi are the lane directions, and σi2 are the corresponding variances which are configurable.

A non-transitory computer-readable medium that stores instructions that, when executed by at least one computing device, will cause the at least one computing device to perform operations comprising:

• obtaining a LiDAR dataset;
• plotting the LiDAR dataset on a 3D graph to define a 3D point cloud;
• using the LiDAR dataset and contents of a vector map to define a cuboid on the 3D graph that encompasses points of the 3D point cloud that are associated with an object in proximity to the vehicle, where the vector map comprises lane information;
• and
• using the cuboid to facilitate driving-related operations of the autonomous vehicle;
• wherein the cuboid is defined on the 3D graph by inferring a heading distribution Ppt(θ) for the object from the 3D point cloud, determining a heading distribution Pm(θ) from the vector map, and performing filter operations to obtain a posterior distribution of a heading using the heading distributions Ppt(θ) and Pm(θ).