Inevitable collision states - a step towards safer robots?
read more
Citations
Planning Algorithms: Introductory Material
Reciprocal n-Body Collision Avoidance
A survey on motion prediction and risk assessment for intelligent vehicles
Sampling-Based Robot Motion Planning: A Review
Algorithms for collision-free navigation of mobile robots in complex cluttered environments: a survey
References
Modern control engineering
Robot Motion Planning
Rapidly-exploring random trees : a new tool for path planning
Randomized kinodynamic planning
The dynamic window approach to collision avoidance
Related Papers (5)
The dynamic window approach to collision avoidance
Frequently Asked Questions (12)
Q2. What have the authors stated for future works in "Inevitable collision states a step towards safer robots?" ?
This paper has introduced the novel concept of inevitable collision states for a given robotic system, ie states for which, no matter what the future trajectory followed by the system is, a collision eventually occurs with an obstacle of the environment. This concept is very general and the authors believe it can be useful both for navigation and motion planning purposes.
Q3. What is the way to avoid unexpected obstacles?
if you believe that there may be unexpected obstacles on the other side, you have two strategies possible:1. Graze the corner while slowing down so that when you pass the corner, your speed is slow enough for you to stop before hitting a possible unexpected obstacle, or2.
Q4. What is the final characterisation of the inevitable collision obstacles?
The final characterisation of the inevitable collision obstacles is determined using:ICOI(B) = ⋂X∈{IS ,I ξT }ICOX (B)which amounts to computing the intersection between a set of generalised polygons.
Q5. What is the Minkowski Sum between A and the control inputs?
IξT is splitinto two subsets Iξ+T and The authorξ− T respectively corresponding to control inputs for which A is accelerating, ie uv ≥ 0, and decelerating, ie uv < 0.
Q6. What is the definition of a robust state?
It characterises robust states as states for which the robotic system can safely stop simply by braking even when placed in a environment with unknown moving obstacles (basically, the field of view is shrunken according to themoving obstacles’ maximum possible speed).
Q7. What is the Minkowski Sum between A and Bi?
when A is turning with the steering angle uξ and decelerating, it eventually crashes into Bi iff it is on a collision course and its distance to Bi is less than d(v).
Q8. What is the Minkowski sum of two sets A and B in a vector space?
The Minkowski sum of two sets A and B in a vector space is equal to {a + b : a ∈ A, b ∈ B} [21].arc of radius b/ tan uξ and arc length d(v) starting from Bi in the −θ direction (Fig. 15 middle).
Q9. What is the concept of inevitable collision states?
On the other hand, assuming that it takes P a certain distance d(v) to slow down and stop, the states corresponding to the wall and the states located at a distance less than d(v) left of the wall are such that, when P is in such a state, no matter what it does in the future, a collision will occur.
Q10. What is the definition of inevitable collision states?
For a start, like its configuration space counterpart, the inevitable collision state concept faces the “curse of dimensionality”, ie the complexity of characterising the inevitable collision states of highdimensional robotic systems.
Q11. What is the simplest way to solve the motion planning problem?
As far as solving the motion planning problem at hand is concerned, it was decided to use a classical motion planning scheme based on the Rapidly-Exploring Random Tree algorithm [23].
Q12. What is the Minkowski Sum3 between B and the segment of length d(v)?
More precisely, it is the Minkowski Sum3 between B and the segment of length d(v) starting from (0, 0) in the −θ direction (Fig. 14 right).