Methods and systems for optimized assignment of traversal tasks under imperfect sensing of autonomous vehicles

A system comprising:

• a memory storing instructions;
• one or more input/output (I/O) interfaces;
• and
• one or more hardware processors coupled to the memory via the one or more I/O interfaces, wherein the one or more hardware processors are configured by the instructions to:
• receive an autonomous vehicle (AV) list having one or more autonomous vehicles (AVs), a task list having one or more traversal tasks to be performed by the one or more AVs, one or more sensors installed in each AV of the AV list, a map of a 3-dimensional (3-D) dynamic environment where each AV to be traversed, and a statistical error model associated with each sensor installed in each AV, wherein each traversal task comprises a predefined source location and a predefined destination location in the 3-D dynamic environment, and the map of the 3-D dynamic environment comprises one or more obstacles and defines one or more environmental attributes associated with each obstacle;
• configure a first radius parameter and a second radius parameter, to define a zone type for each AV with respect to each obstacle, from a zone type list having (i) a red zone, (ii) a yellow zone, and (iii) a green zone, based on a distance from each obstacle to each AV;
• determine (i) an estimated completion time, and (ii) an estimated collision count, for each traversal task with respect to each AV, using (i) the statistical error model associated with each sensor installed in the associated AV, (ii) the one or more obstacles present during completion of associated traversal task, and (iii) the one or more environmental attributes associated with each obstacle;
• identify and assign, via the one or more hardware processors, a traversal task among the one or more traversal tasks to each AV, based the estimated completion time and the estimated collision count being lowest, and removing each traversal task that is assigned from the task list;
• determine a total completion time, for each AV, as a summation of the estimated completion time of the each of the one or more traversal tasks assigned to the associated AV;
• assign the traversal task that is unassigned and present in the task list, to the AV having the total completion time and the estimated collision count being the lowest, and removing each traversal task that is assigned from the task list;
• and
• perform (i) determining the total completion time for each AV, (ii) assigning the traversal task that is unassigned and present in the task list, to the AV having the total completion time and the estimated collision count being the lowest, and (iii) removing each traversal task that is assigned from the task list, until either each of the one or more traversal tasks are assigned or until the task list is empty.

One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:

• receiving, an autonomous vehicle (AV) list having one or more autonomous vehicles (AVs), a task list having one or more traversal tasks to be performed by the one or more AVs, one or more sensors installed in each AV of the AV list, a map of a 3-dimensional dynamic environment where each AV to be traversed, and a statistical error model associated with each sensor installed in each AV, wherein each traversal task comprises a predefined source location and a predefined destination location in the 3-D dynamic environment, and the map of the 3-D dynamic environment comprises one or more obstacles and defines one or more environmental attributes associated with each obstacle;
• configuring, a first radius parameter and a second radius parameter, to define a zone type for each AV with respect to each obstacle, from a zone type list having (i) a red zone, (ii) a yellow zone, and (iii) a green zone, based on a distance from each obstacle to each AV;
• determining, (i) an estimated completion time, and (ii) an estimated collision count, for each traversal task with respect to each AV, using (i) the statistical error model associated with each sensor installed in the associated AV, (ii) the one or more obstacles present during completion of associated traversal task, and (iii) the one or more environmental attributes associated with each obstacle;
• identifying and assigning, a traversal task among the one or more traversal tasks to each AV, based the estimated completion time and the estimated collision count being lowest, and removing each traversal task that is assigned from the task list;
• determining, a total completion time, for each AV, as a summation of the estimated completion time of the each of the one or more traversal tasks assigned to the associated AV;
• assigning, the traversal task that is unassigned and present in the task list, to the AV having the total completion time and the estimated collision count being the lowest, and removing each traversal task that is assigned from the task list;
• and
• performing, (i) determining the total completion time for each AV, (ii) assigning the traversal task that is unassigned and present in the task list, to the AV having the total completion time and the estimated collision count being the lowest, and (iii) removing each traversal task that is assigned from the task list, until either each of the one or more traversal tasks are assigned or until the task list is empty.