17900258. TRAJECTORY OPTIMIZATION IN MULTI-AGENT ENVIRONMENTS simplified abstract (Zoox, Inc.)
Contents
- 1 TRAJECTORY OPTIMIZATION IN MULTI-AGENT ENVIRONMENTS
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 TRAJECTORY OPTIMIZATION IN MULTI-AGENT ENVIRONMENTS - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
TRAJECTORY OPTIMIZATION IN MULTI-AGENT ENVIRONMENTS
Organization Name
Inventor(s)
Marin Kobilarov of Baltimore MD (US)
Chonhyon Park of San Jose CA (US)
TRAJECTORY OPTIMIZATION IN MULTI-AGENT ENVIRONMENTS - A simplified explanation of the abstract
This abstract first appeared for US patent application 17900258 titled 'TRAJECTORY OPTIMIZATION IN MULTI-AGENT ENVIRONMENTS
Simplified Explanation
The patent application discusses techniques for determining optimal driving trajectories for autonomous vehicles in complex multi-agent driving environments. A baseline trajectory is perturbed and parameterized into a vector of vehicle states associated with different segments of the trajectory. This vector is modified to ensure the resultant perturbed trajectory is kino-dynamically feasible. The perturbed trajectory is input into a prediction model trained to output a predicted future driving scene, including predicted future states for the vehicle and trajectories for additional agents in the environment. Costs associated with each perturbed trajectory are evaluated to determine the optimal trajectory for controlling the vehicle in the driving environment.
- Determining optimal driving trajectories for autonomous vehicles in complex multi-agent driving environments
- Perturbing and parameterizing a baseline trajectory into a vector of vehicle states
- Modifying the vector to ensure kino-dynamic feasibility
- Inputting the perturbed trajectory into a prediction model to output a predicted future driving scene
- Evaluating costs associated with each perturbed trajectory to determine the optimal trajectory
Potential Applications
This technology could be applied in autonomous vehicles, transportation systems, and robotics for efficient and safe navigation in complex environments.
Problems Solved
This technology solves the problem of determining optimal driving trajectories in dynamic and complex driving environments, ensuring safe and efficient navigation for autonomous vehicles.
Benefits
The benefits of this technology include improved safety, efficiency, and adaptability in autonomous vehicle navigation, leading to reduced accidents and smoother traffic flow.
Potential Commercial Applications
Potential commercial applications of this technology include autonomous vehicle systems, transportation logistics, and robotic systems for various industries.
Possible Prior Art
One possible prior art for this technology could be related to trajectory optimization algorithms used in autonomous vehicles and robotics for path planning and navigation in dynamic environments.
What are the specific prediction models used in this technology?
The specific prediction models used in this technology are not explicitly mentioned in the abstract. Further details on the types of prediction models and algorithms employed would provide a deeper understanding of the predictive capabilities of the system.
How are the costs associated with each perturbed trajectory evaluated in the optimization algorithm?
The abstract mentions that costs associated with each perturbed trajectory are evaluated to determine the optimal trajectory. Understanding the specific cost functions and criteria used in the evaluation process would shed light on the decision-making process of the optimization algorithm.
Original Abstract Submitted
Techniques are discussed herein for determining optimal driving trajectories for autonomous vehicles in complex multi-agent driving environments. A baseline trajectory may be perturbed and parameterized into a vector of vehicle states associated with different segments (or portions) of the trajectory. Such a vector may be modified to ensure the resultant perturbed trajectory is kino-dynamically feasible. The vectorized perturbed trajectory may be input, including a representation of the current driving environment and additional agents, into a prediction model trained to output a predicted future driving scene. The predicted future driving scene, including predicted future states for the vehicle and predicted trajectories for the additional agents in the environment, may be evaluated to determine costs associated with each perturbed trajectory. Based on the determined costs, the optimization algorithm may determine subsequent perturbations and/or the optimal trajectory for controlling the vehicle in the driving environment.
