Vehicle autonomous collision prediction and escaping system (ACE)

A method, comprising:

• sensing, by a sensor of a vehicle, sensor information relating to an environment surrounding the vehicle;
• sending, by the sensor, the sensor information to a processor;
• identifying, by the processor, an object within the environment surrounding the vehicle based on the sensor information;
• determining, by the processor, a probabilistic vector for the object based on a stochastic model;
• determining, by the processor based on the probabilistic vector for the object, if the object may collide with the vehicle;
• determining, by the processor, a best path to avoid the object from a tree comprising potential paths for the vehicle generated by the processor determining a root node for the tree, the root node being associated with a current position of the vehicle at a first time, wherein determining the best path includes:
• determining two or more child nodes depending from the root node in the tree and corresponding to two or more different possible positions for the vehicle at a second time subsequent to the first time, determining probabilities of collision between the object and the vehicle for the two or more child nodes, determining weights for the two or more child nodes, normalizing the two or more child nodes by multiplying a respective weight by a respective probability of collision for each of the two or more child nodes, determining, prior to the second time and based on the probabilistic vector for the object and the two or more different possible positions for the vehicle at the second time, a single child node of the two or more child nodes having a lowest probability of a collision between the object and the vehicle at the second time, determining two or more grandchild nodes depending from the single child node in the tree and corresponding to two or more different possible positions for the vehicle at a third time subsequent to the second time, and determining, prior to the third time and based on the probabilistic vector for the object and the two or more different possible positions for the vehicle at the third time, a single grandchild node of the two or more grandchild nodes having a lowest probability of a collision between the object and the vehicle at the third time, and wherein the best path to avoid the object is one of the potential paths linking the root node to the single grandchild node via the single child node in the tree;
• providing, by the processor, the best path to a vehicle control system;
• and
• controlling, autonomously by the vehicle control system, driving functions of the vehicle to maneuver the vehicle along the best path and avoid collision with the object.

A non-transitory information storage media having stored thereon one or more instructions, that when executed by one or more processors, cause a vehicle to perform a method, the method comprising:

• sensing an environment surrounding the vehicle to obtain sensor information;
• identifying an object within the environment based on the sensor information;
• determining a probabilistic vector for the object based on a stochastic model, wherein the probabilistic vector for the object is modified by a constraint, wherein the constraint is based on a law of physics, a rule, a law, a machine-learned behavior, and/or an ability of the object;
• determining, based on the probabilistic vector for the object, if the object may collide with the vehicle;
• generating, in response to determining that the object may collide with the vehicle, a tree comprising potential paths for the vehicle to follow relative to the object, wherein the generating the tree comprises:
• determining a root node for the tree, the root node being associated with a current position of the vehicle at a first time;
• determining two or more child nodes depending from the root node in the tree and corresponding to two or more different possible positions for the vehicle at a second time subsequent to the first time;
• determining probabilities of collision between the object and the vehicle for the two or more child nodes;
• determining weights for the two or more child nodes;
• normalizing the two or more child nodes by multiplying a respective weight by a respective probability of collision for each of the two or more child nodes;
• determining, prior to the second time and based on the probabilistic vector for the object and the two or more different possible positions for the vehicle at the second time, a single child node of the two or more child nodes having a lowest probability of a collision between the object and the vehicle at the second time;
• determining two or more grandchild nodes depending from the single child node in the tree and corresponding to two or more different possible positions for the vehicle at a third time subsequent to the second time;
• and
• determining, prior to the third time and based on the probabilistic vector for the object and the two or more different possible positions for the vehicle at the third time, a single grandchild node of the two or more grandchild nodes having a lowest probability of a collision between the object and the vehicle at the third time;
• determining a best path for the vehicle to follow to avoid the object by traversing the tree, wherein traversing the tree comprises evaluating dependent nodes from the root node in the tree only from a corresponding parent node having a lowest probability of a collision at a corresponding time, and wherein the best path is a path from the root node to the single grandchild node via the single child node in the tree;
• providing the best path to a vehicle control system;
• and
• controlling, autonomously by the vehicle control system, driving functions of the vehicle to maneuver the vehicle along the best path and avoid collision with the object.